From 403ff67c3018710ca63b076a30ae9c8445923a68 Mon Sep 17 00:00:00 2001 From: DanielleWashington Date: Fri, 7 Aug 2026 09:38:18 -0400 Subject: [PATCH] docs: expose Agent endpoints --- advanced-scraping-guide.mdx | 2 +- api-reference/endpoint/agent-delete.mdx | 4 +--- api-reference/endpoint/agent-get.mdx | 4 +--- api-reference/endpoint/agent.mdx | 2 -- api-reference/v2-openapi.json | 4 ++-- .../choosing-the-data-extractor.mdx | 2 +- docs.json | 8 +------- es/advanced-scraping-guide.mdx | 4 ++-- es/api-reference/endpoint/agent-delete.mdx | 4 +--- es/api-reference/endpoint/agent-get.mdx | 4 +--- es/api-reference/endpoint/agent.mdx | 4 +--- es/api-reference/v2-openapi.json | 6 +++--- .../choosing-the-data-extractor.mdx | 4 ++-- es/features/agent.mdx | 19 ++++++++++--------- es/features/models.mdx | 10 +++++----- es/sdks/cli.mdx | 4 ++-- features/agent.mdx | 19 +++++++++++-------- features/models.mdx | 10 ++++++---- fr/advanced-scraping-guide.mdx | 4 ++-- fr/api-reference/endpoint/agent-delete.mdx | 4 +--- fr/api-reference/endpoint/agent-get.mdx | 4 +--- fr/api-reference/endpoint/agent.mdx | 4 +--- fr/api-reference/v2-openapi.json | 6 +++--- .../choosing-the-data-extractor.mdx | 4 ++-- fr/features/agent.mdx | 19 ++++++++++--------- fr/features/models.mdx | 10 +++++----- fr/sdks/cli.mdx | 4 ++-- ja/advanced-scraping-guide.mdx | 4 ++-- ja/api-reference/endpoint/agent-delete.mdx | 4 +--- ja/api-reference/endpoint/agent-get.mdx | 4 +--- ja/api-reference/endpoint/agent.mdx | 4 +--- ja/api-reference/v2-openapi.json | 6 +++--- .../choosing-the-data-extractor.mdx | 4 ++-- ja/features/agent.mdx | 19 ++++++++++--------- ja/features/models.mdx | 10 +++++----- ja/sdks/cli.mdx | 4 ++-- pt-BR/advanced-scraping-guide.mdx | 4 ++-- pt-BR/api-reference/endpoint/agent-delete.mdx | 4 +--- pt-BR/api-reference/endpoint/agent-get.mdx | 4 +--- pt-BR/api-reference/endpoint/agent.mdx | 4 +--- pt-BR/api-reference/v2-openapi.json | 6 +++--- .../choosing-the-data-extractor.mdx | 4 ++-- pt-BR/features/agent.mdx | 19 ++++++++++--------- pt-BR/features/models.mdx | 10 +++++----- pt-BR/sdks/cli.mdx | 4 ++-- sdks/cli.mdx | 2 +- snippets/es/v2/agent/with-model/curl.mdx | 4 ++-- snippets/es/v2/agent/with-model/js.mdx | 6 +++--- snippets/es/v2/agent/with-model/python.mdx | 4 ++-- snippets/fr/v2/agent/with-model/curl.mdx | 4 ++-- snippets/fr/v2/agent/with-model/js.mdx | 6 +++--- snippets/fr/v2/agent/with-model/python.mdx | 4 ++-- snippets/ja/v2/agent/with-model/curl.mdx | 4 ++-- snippets/ja/v2/agent/with-model/js.mdx | 6 +++--- snippets/ja/v2/agent/with-model/python.mdx | 4 ++-- snippets/pt-BR/v2/agent/with-model/curl.mdx | 4 ++-- snippets/pt-BR/v2/agent/with-model/js.mdx | 6 +++--- snippets/pt-BR/v2/agent/with-model/python.mdx | 4 ++-- snippets/v2/agent/with-model/curl.mdx | 5 ++--- snippets/v2/agent/with-model/js.mdx | 5 ++--- snippets/v2/agent/with-model/python.mdx | 5 ++--- snippets/zh/v2/agent/with-model/curl.mdx | 4 ++-- snippets/zh/v2/agent/with-model/js.mdx | 6 +++--- snippets/zh/v2/agent/with-model/python.mdx | 4 ++-- zh/advanced-scraping-guide.mdx | 4 ++-- zh/api-reference/endpoint/agent-delete.mdx | 4 +--- zh/api-reference/endpoint/agent-get.mdx | 4 +--- zh/api-reference/endpoint/agent.mdx | 4 +--- zh/api-reference/v2-openapi.json | 6 +++--- .../choosing-the-data-extractor.mdx | 4 ++-- zh/features/agent.mdx | 19 ++++++++++--------- zh/features/models.mdx | 10 +++++----- zh/sdks/cli.mdx | 4 ++-- 73 files changed, 201 insertions(+), 236 deletions(-) diff --git a/advanced-scraping-guide.mdx b/advanced-scraping-guide.mdx index e5b2c917a..61aa3fc8a 100644 --- a/advanced-scraping-guide.mdx +++ b/advanced-scraping-guide.mdx @@ -464,7 +464,7 @@ Use the `/v2/agent` endpoint for autonomous, multi-page data extraction. The age | `schema` | `object` | — | JSON schema to structure the extracted data. | | `maxCredits` | `number` | `2500` | Maximum credits the agent can spend. The dashboard supports up to 2,500; for higher limits, set this via the API (values above 2,500 are always billed as paid requests). | | `strictConstrainToURLs` | `boolean` | `false` | When `true`, the agent only visits the provided URLs. | -| `model` | `string` | `"spark-1-mini"` | AI model to use. `"spark-1-mini"` (default, 60% cheaper) or `"spark-1-pro"` (higher accuracy). | +| `model` | `string` | `"spark-1-pro"` | AI model to use. `"spark-1-pro"` (default, higher accuracy) or `"spark-1-mini"` (60% cheaper). | diff --git a/api-reference/endpoint/agent-delete.mdx b/api-reference/endpoint/agent-delete.mdx index b6984c59f..c7c21ff7a 100644 --- a/api-reference/endpoint/agent-delete.mdx +++ b/api-reference/endpoint/agent-delete.mdx @@ -1,9 +1,7 @@ --- title: 'Cancel Agent' openapi: '/api-reference/v2-openapi.json DELETE /agent/{jobId}' -hidden: true -noindex: true --- -> Are you an AI agent that needs a Firecrawl API key? See [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) for automated onboarding instructions. \ No newline at end of file +> Are you an AI agent that needs a Firecrawl API key? See [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) for automated onboarding instructions. diff --git a/api-reference/endpoint/agent-get.mdx b/api-reference/endpoint/agent-get.mdx index 7bdaef331..4eb647895 100644 --- a/api-reference/endpoint/agent-get.mdx +++ b/api-reference/endpoint/agent-get.mdx @@ -1,9 +1,7 @@ --- title: 'Get Agent Status' openapi: '/api-reference/v2-openapi.json GET /agent/{jobId}' -hidden: true -noindex: true --- -> Are you an AI agent that needs a Firecrawl API key? See [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) for automated onboarding instructions. \ No newline at end of file +> Are you an AI agent that needs a Firecrawl API key? See [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) for automated onboarding instructions. diff --git a/api-reference/endpoint/agent.mdx b/api-reference/endpoint/agent.mdx index 987860b91..7c1d9fbc9 100644 --- a/api-reference/endpoint/agent.mdx +++ b/api-reference/endpoint/agent.mdx @@ -1,8 +1,6 @@ --- title: 'Agent' openapi: '/api-reference/v2-openapi.json POST /agent' -hidden: true -noindex: true --- diff --git a/api-reference/v2-openapi.json b/api-reference/v2-openapi.json index 32c09bb26..f8c706fc2 100644 --- a/api-reference/v2-openapi.json +++ b/api-reference/v2-openapi.json @@ -2513,8 +2513,8 @@ "spark-1-mini", "spark-1-pro" ], - "default": "spark-1-mini", - "description": "The model to use for the agent task. spark-1-mini (default) is 60% cheaper, spark-1-pro offers higher accuracy for complex tasks" + "default": "spark-1-pro", + "description": "The model to use for the agent task. spark-1-pro is the default and offers higher accuracy for complex tasks; spark-1-mini is 60% cheaper" }, "auditMetadata": { "$ref": "#/components/schemas/AuditMetadata" diff --git a/developer-guides/usage-guides/choosing-the-data-extractor.mdx b/developer-guides/usage-guides/choosing-the-data-extractor.mdx index 29f69e747..e4527b2b8 100644 --- a/developer-guides/usage-guides/choosing-the-data-extractor.mdx +++ b/developer-guides/usage-guides/choosing-the-data-extractor.mdx @@ -41,7 +41,7 @@ The `/agent` endpoint is Firecrawl's most advanced offering—the successor to ` - **Autonomous Navigation**: The agent searches and navigates deep into sites to find your data - **Deep Web Search**: Autonomously discovers information across multiple domains and pages - **Parallel Processing**: Processes multiple sources simultaneously for faster results -- **Models Available**: `spark-1-mini` (default, 60% cheaper) and `spark-1-pro` (higher accuracy) +- **Models Available**: `spark-1-pro` (default, higher accuracy) and `spark-1-mini` (60% cheaper) ### Example diff --git a/docs.json b/docs.json index 5b581cb39..9ff3b5819 100755 --- a/docs.json +++ b/docs.json @@ -522,7 +522,6 @@ }, { "group": "Agent Endpoints", - "hidden": true, "pages": [ "api-reference/endpoint/agent", "api-reference/endpoint/agent-get", @@ -1320,7 +1319,6 @@ }, { "group": "Endpoints de agentes", - "hidden": true, "pages": [ "es/api-reference/endpoint/agent", "es/api-reference/endpoint/agent-get", @@ -2118,7 +2116,6 @@ }, { "group": "Points de terminaison Agent", - "hidden": true, "pages": [ "fr/api-reference/endpoint/agent", "fr/api-reference/endpoint/agent-get", @@ -2916,7 +2913,6 @@ }, { "group": "エージェントエンドポイント", - "hidden": true, "pages": [ "ja/api-reference/endpoint/agent", "ja/api-reference/endpoint/agent-get", @@ -3714,7 +3710,6 @@ }, { "group": "Endpoints de agente", - "hidden": true, "pages": [ "pt-BR/api-reference/endpoint/agent", "pt-BR/api-reference/endpoint/agent-get", @@ -4512,7 +4507,6 @@ }, { "group": "代理端点", - "hidden": true, "pages": [ "zh/api-reference/endpoint/agent", "zh/api-reference/endpoint/agent-get", @@ -5340,4 +5334,4 @@ } }, "theme": "aspen" -} \ No newline at end of file +} diff --git a/es/advanced-scraping-guide.mdx b/es/advanced-scraping-guide.mdx index 5d6d74d2a..47efee92f 100644 --- a/es/advanced-scraping-guide.mdx +++ b/es/advanced-scraping-guide.mdx @@ -486,7 +486,7 @@ Usa el endpoint `/v2/agent` para la extracción autónoma de datos en múltiples | `schema` | `object` | — | Esquema JSON para estructurar los datos extraídos. | | `maxCredits` | `number` | `2500` | Máximo de créditos que el agente puede consumir. El panel de control admite hasta 2.500; para límites superiores, configúrelo mediante la API (los valores por encima de 2.500 siempre se facturan como solicitudes de pago). | | `strictConstrainToURLs` | `boolean` | `false` | Cuando su valor es `true`, el agente solo visita las URLs proporcionadas. | -| `model` | `string` | `"spark-1-mini"` | Modelo de IA a utilizar: `"spark-1-mini"` (predeterminado, 60% más económico) o `"spark-1-pro"` (mayor precisión). | +| `model` | `string` | `"spark-1-pro"` | Modelo de IA a utilizar: `"spark-1-pro"` (predeterminado, mayor precisión) o `"spark-1-mini"` (60% más económico). | ```python Python @@ -717,4 +717,4 @@ Si tu firewall bloquea las solicitudes salientes de tu aplicación hacia servici - **Dirección IP**: `35.245.250.27` -Agrega esta IP a la lista de permitidos de salida de tu firewall para que tu backend pueda enviar solicitudes de scraping, crawling, mapping y de agentes a Firecrawl. \ No newline at end of file +Agrega esta IP a la lista de permitidos de salida de tu firewall para que tu backend pueda enviar solicitudes de scraping, crawling, mapping y de agentes a Firecrawl. diff --git a/es/api-reference/endpoint/agent-delete.mdx b/es/api-reference/endpoint/agent-delete.mdx index 58a271db2..27999a879 100644 --- a/es/api-reference/endpoint/agent-delete.mdx +++ b/es/api-reference/endpoint/agent-delete.mdx @@ -1,8 +1,6 @@ --- title: 'Cancelar agente' openapi: '/es/api-reference/v2-openapi.json DELETE /agent/{jobId}' -hidden: true -noindex: true --- -> ¿Eres un agente de IA que necesita una clave de API de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automatizada. \ No newline at end of file +> ¿Eres un agente de IA que necesita una clave de API de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automatizada. diff --git a/es/api-reference/endpoint/agent-get.mdx b/es/api-reference/endpoint/agent-get.mdx index 494d11a24..d2a04ab70 100644 --- a/es/api-reference/endpoint/agent-get.mdx +++ b/es/api-reference/endpoint/agent-get.mdx @@ -1,8 +1,6 @@ --- title: 'Obtener el estado del agente' openapi: '/es/api-reference/v2-openapi.json GET /agent/{jobId}' -hidden: true -noindex: true --- -> ¿Eres un agente de IA que necesita una API key de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automática. \ No newline at end of file +> ¿Eres un agente de IA que necesita una API key de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automática. diff --git a/es/api-reference/endpoint/agent.mdx b/es/api-reference/endpoint/agent.mdx index efdc406f0..dcded7f46 100644 --- a/es/api-reference/endpoint/agent.mdx +++ b/es/api-reference/endpoint/agent.mdx @@ -1,8 +1,6 @@ --- title: 'Agente' openapi: '/es/api-reference/v2-openapi.json POST /agent' -hidden: true -noindex: true --- -> ¿Eres un agente de IA que necesita una clave de API de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para ver instrucciones de incorporación automatizada. \ No newline at end of file +> ¿Eres un agente de IA que necesita una clave de API de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para ver instrucciones de incorporación automatizada. diff --git a/es/api-reference/v2-openapi.json b/es/api-reference/v2-openapi.json index d537af625..bd570c59a 100644 --- a/es/api-reference/v2-openapi.json +++ b/es/api-reference/v2-openapi.json @@ -4128,8 +4128,8 @@ "type": "number" }, "model": { - "default": "spark-1-mini", - "description": "El modelo que utilizará el agente para la tarea. spark-1-mini (predeterminado) es un 60 % más barato; spark-1-pro ofrece mayor precisión para tareas complejas.", + "default": "spark-1-pro", + "description": "El modelo que utilizará el agente para la tarea. spark-1-pro es el predeterminado y ofrece mayor precisión para tareas complejas; spark-1-mini es un 60 % más barato.", "enum": [ "spark-1-mini", "spark-1-pro" @@ -9906,4 +9906,4 @@ "url": "https://api.firecrawl.dev/v2" } ] -} \ No newline at end of file +} diff --git a/es/developer-guides/usage-guides/choosing-the-data-extractor.mdx b/es/developer-guides/usage-guides/choosing-the-data-extractor.mdx index fb792b4b2..26b92a9f6 100644 --- a/es/developer-guides/usage-guides/choosing-the-data-extractor.mdx +++ b/es/developer-guides/usage-guides/choosing-the-data-extractor.mdx @@ -47,7 +47,7 @@ El endpoint `/agent` es la funcionalidad más avanzada de Firecrawl, el sucesor * **Navegación autónoma**: El agente busca y navega en profundidad por sitios web para encontrar tus datos * **Búsqueda profunda en la web**: Descubre información de forma autónoma en múltiples dominios y páginas * **Procesamiento en paralelo**: Procesa múltiples fuentes simultáneamente para obtener resultados más rápidos -* **Modelos disponibles**: `spark-1-mini` (predeterminado, 60% más económico) y `spark-1-pro` (mayor precisión) +* **Modelos disponibles**: `spark-1-pro` (predeterminado, mayor precisión) y `spark-1-mini` (60% más económico)
### Ejemplo @@ -272,4 +272,4 @@ La ventaja clave es que, con `/agent`, puedes prescindir por completo de las URL * [Modelos de Agent](/es/features/models) * [Documentación del modo JSON](/es/features/llm-extract) * [Documentación de Extract](/es/features/extract) -* [Rastreo por lotes](/es/features/batch-scrape) \ No newline at end of file +* [Rastreo por lotes](/es/features/batch-scrape) diff --git a/es/features/agent.mdx b/es/features/agent.mdx index 2d24db0e6..5215ab46f 100644 --- a/es/features/agent.mdx +++ b/es/features/agent.mdx @@ -5,7 +5,6 @@ og:title: "Agent | Firecrawl" og:description: "Recopila datos dondequiera que se encuentren en la web. Describe lo que quieres, /agent se encarga del resto." sidebarTitle: "Agent" icon: "magic" -noindex: true --- @@ -143,19 +142,19 @@ Puedes compartir ejecuciones del agente directamente desde el Agent Playground. ## Selección de modelo
-Firecrawl Agent ofrece dos modelos. **Spark 1 Mini es un 60% más barato** y es el modelo predeterminado — perfecto para la mayoría de los casos de uso. Pásate a Spark 1 Pro cuando necesites la máxima precisión en tareas complejas. +Firecrawl Agent ofrece dos modelos. `spark-1-pro` es el modelo predeterminado y ofrece la máxima precisión para tareas complejas. `spark-1-mini` cuesta un 60% menos y es una buena opción cuando priorizas la eficiencia de costes. | Modelo | Coste | Precisión | Ideal para | |--------|-------|-----------|------------| -| `spark-1-mini` | **60% más barato** | Estándar | La mayoría de las tareas (predeterminado) | -| `spark-1-pro` | Estándar | Mayor | Investigación compleja, extracción crítica | +| `spark-1-mini` | **60% más barato** | Estándar | Tareas con prioridad en el coste | +| `spark-1-pro` | Estándar | Mayor | Investigación compleja, extracción crítica (predeterminado) | - **Empieza con Spark 1 Mini** (predeterminado) — gestiona bien la mayoría de las tareas de extracción con un coste un 60% menor. Cambia a Pro solo para investigación compleja en múltiples dominios o cuando la precisión sea crítica. + Agent usa Spark 1 Pro de forma predeterminada. Elige Spark 1 Mini explícitamente cuando quieras reducir el coste de tareas sencillas.
- ### Spark 1 Mini (predeterminado) + ### Spark 1 Mini
`spark-1-mini` es nuestro modelo eficiente, ideal para tareas sencillas de extracción de datos. @@ -201,7 +200,7 @@ Pasa el parámetro `model` para seleccionar el modelo que quieres usar: | Parámetro | Tipo | Obligatorio | Descripción | | ------------ | ------ | ----------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | `prompt` | string | **Sí** | Descripción en lenguaje natural de los datos que deseas extraer (máx. 10.000 caracteres) | -| `model` | string | No | Modelo a utilizar: `spark-1-mini` (predeterminado) o `spark-1-pro` | +| `model` | string | No | Modelo a utilizar: `spark-1-pro` (predeterminado) o `spark-1-mini` | | `urls` | array | No | Lista opcional de URLs en las que centrar la extracción | | `schema` | object | No | Esquema JSON opcional para una salida estructurada | | `maxCredits` | number | No | Número máximo de créditos que se pueden usar en esta tarea del agente. De forma predeterminada es **2,500** si no se especifica. El panel de control admite valores de hasta **2,500**; para límites superiores, configura `maxCredits` mediante la API (los valores por encima de 2,500 siempre se tratan como solicitudes de pago). Si se alcanza el límite, el trabajo falla y **no se devuelve ningún dato**. Las ejecuciones fallidas no se facturan: los créditos usados para el razonamiento de IA nunca se cobran en caso de fallo, cualquier crédito usado para llamadas a herramientas durante la ejecución (`scraping`, `search`, `mapping`, etc.) se reembolsa, y la respuesta informa `creditsUsed: 0`. | @@ -256,7 +255,9 @@ Consulta la [documentación de Ask](/es/features/ask) para conocer todos los det ## Referencia de la API -Consulta la [Agent API Reference](/es/api-reference/endpoint/agent) para más detalles. +- [Iniciar una tarea de Agent](/es/api-reference/endpoint/agent) +- [Obtener el estado de Agent](/es/api-reference/endpoint/agent-get) +- [Cancelar una tarea de Agent](/es/api-reference/endpoint/agent-delete) ¿Tienes comentarios o necesitas ayuda? Escríbenos a [help@firecrawl.com](mailto:help@firecrawl.com). @@ -314,4 +315,4 @@ Prueba Agent ahora en [firecrawl.dev/app/agent](https://www.firecrawl.dev/app/ag Los precios están sujetos a cambios a medida que pasamos de la Research Preview a la disponibilidad general. Los usuarios actuales recibirán aviso previo de cualquier actualización de precios. -> ¿Eres un Agent de IA que necesita una API key de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automatizada. \ No newline at end of file +> ¿Eres un Agent de IA que necesita una API key de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automatizada. diff --git a/es/features/models.mdx b/es/features/models.mdx index 6ef48c258..ec72ec071 100644 --- a/es/features/models.mdx +++ b/es/features/models.mdx @@ -21,15 +21,15 @@ Firecrawl Agent ofrece dos modelos optimizados para distintos casos de uso. Elig | Model | Cost | Accuracy | Best For | |-------|------|----------|----------| -| `spark-1-mini` | **60% más económico** | Estándar | La mayoría de las tareas (predeterminado) | -| `spark-1-pro` | Estándar | Más alta | Investigación compleja, extracción crítica | +| `spark-1-mini` | **60% más económico** | Estándar | Tareas con prioridad en el coste | +| `spark-1-pro` | Estándar | Más alta | Investigación compleja, extracción crítica (predeterminado) | - **Empieza con Spark 1 Mini** (predeterminado): gestiona bien la mayoría de las tareas de extracción con un coste un 60% menor. Cambia a Pro solo para investigación compleja en múltiples dominios o cuando la precisión sea crítica. + Agent usa Spark 1 Pro de forma predeterminada. Elige Spark 1 Mini explícitamente cuando quieras reducir el coste de tareas sencillas.
- ## Spark 1 Mini (por defecto) + ## Spark 1 Mini
`spark-1-mini` es nuestro modelo eficiente, ideal para tareas sencillas de extracción de datos. @@ -139,4 +139,4 @@ Consulta la [Referencia de la API del agente](/es/api-reference/endpoint/agent) ¿Tienes preguntas sobre qué modelo usar? Envía un correo a [help@firecrawl.com](mailto:help@firecrawl.com). -> ¿Eres un agente de IA que necesita una clave de API de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automatizadas. \ No newline at end of file +> ¿Eres un agente de IA que necesita una clave de API de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automatizadas. diff --git a/es/sdks/cli.mdx b/es/sdks/cli.mdx index ff94990e4..1ea535a5d 100644 --- a/es/sdks/cli.mdx +++ b/es/sdks/cli.mdx @@ -399,7 +399,7 @@ Busca y recopila datos de la web usando indicaciones en lenguaje natural. | Opción | Descripción | | --------------------------- | ---------------------------------------------------------------------------------------------------- | | `--urls ` | Lista opcional de URL en las que enfocar el agente (separadas por comas) | -| `--model ` | Modelo a utilizar: `spark-1-mini` (predeterminado, 60% más barato) o `spark-1-pro` (mayor precisión) | +| `--model ` | Modelo a utilizar: `spark-1-pro` (predeterminado, mayor precisión) o `spark-1-mini` (60% más barato) | | `--schema ` | Esquema JSON para salida estructurada (cadena JSON en línea) | | `--schema-file ` | Ruta al archivo de esquema JSON para salida estructurada | | `--max-credits ` | Créditos máximos que se pueden gastar (el trabajo falla si se alcanza el límite) | @@ -588,4 +588,4 @@ La CLI de Firecrawl y los tres segmentos de skills son de código abierto en Git * [`firecrawl/skills`](https://github.com/firecrawl/skills) — skills de Build (integrar Firecrawl en el código de la aplicación) * [`firecrawl/firecrawl-workflows`](https://github.com/firecrawl/firecrawl-workflows) — skills de flujo de trabajo (entregables repetibles, como informes de investigación, auditorías SEO, listas de leads y clones de diseño) -> ¿Eres un agente de IA que necesita una clave de API de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automatizada. \ No newline at end of file +> ¿Eres un agente de IA que necesita una clave de API de Firecrawl? Consulta [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para obtener instrucciones de incorporación automatizada. diff --git a/features/agent.mdx b/features/agent.mdx index e72bec031..5ae284c2c 100644 --- a/features/agent.mdx +++ b/features/agent.mdx @@ -5,7 +5,6 @@ og:title: "Agent | Firecrawl" og:description: "Gather data wherever it lives on the web. Describe what you want, /agent handles the rest." sidebarTitle: "Agent" icon: "magic" -noindex: true --- @@ -128,18 +127,20 @@ You can share agent runs directly from the Agent playground. Shared links are pu ## Model Selection -Firecrawl Agent offers two models. **Spark 1 Mini is 60% cheaper** and is the default — perfect for most use cases. Upgrade to Spark 1 Pro when you need maximum accuracy on complex tasks. +Firecrawl Agent offers two models. `spark-1-pro` is the default and offers the highest accuracy for complex tasks. `spark-1-mini` costs 60% less and is a good fit when cost efficiency matters. | Model | Cost | Accuracy | Best For | |-------|------|----------|----------| -| `spark-1-mini` | **60% cheaper** | Standard | Most tasks (default) | -| `spark-1-pro` | Standard | Higher | Complex research, critical extraction | +| `spark-1-mini` | **60% cheaper** | Standard | Cost-efficient tasks | +| `spark-1-pro` | Standard | Higher | Complex research, critical extraction (default) | -**Start with Spark 1 Mini** (default) — it handles most extraction tasks well at 60% lower cost. Switch to Pro only for complex multi-domain research or when accuracy is critical. +Agent uses Spark 1 Pro by default. Choose Spark 1 Mini explicitly when you want to reduce cost for straightforward tasks. -### Spark 1 Mini (Default) +
+ ### Spark 1 Mini +
`spark-1-mini` is our efficient model, ideal for straightforward data extraction tasks. @@ -176,7 +177,7 @@ Pass the `model` parameter to select which model to use: | Parameter | Type | Required | Description | |-----------|------|----------|-------------| | `prompt` | string | **Yes** | Natural language description of the data you want to extract (max 10,000 characters) | -| `model` | string | No | Model to use: `spark-1-mini` (default) or `spark-1-pro` | +| `model` | string | No | Model to use: `spark-1-pro` (default) or `spark-1-mini` | | `urls` | array | No | Optional list of URLs to focus the extraction | | `schema` | object | No | Optional JSON schema for structured output | | `maxCredits` | number | No | Maximum number of credits to spend on this agent task. Defaults to **2,500** if not set. The dashboard supports values up to **2,500**; for higher limits, set `maxCredits` via the API (values above 2,500 are always treated as paid requests). If the limit is reached, the job fails and **no data is returned**. Failed runs are not billed: credits used for AI reasoning are never charged on failure, any credits used for tool calls during the run (scraping, search, mapping, etc.) are refunded, and the response reports `creditsUsed: 0`. | @@ -221,7 +222,9 @@ See the [Ask documentation](/features/ask) for full details and integration exam ## API Reference -Check out the [Agent API Reference](/api-reference/endpoint/agent) for more details. +- [Start an Agent task](/api-reference/endpoint/agent) +- [Get Agent status](/api-reference/endpoint/agent-get) +- [Cancel an Agent task](/api-reference/endpoint/agent-delete) Have feedback or need help? Email [help@firecrawl.com](mailto:help@firecrawl.com). diff --git a/features/models.mdx b/features/models.mdx index bb4a580c9..f75b2b71a 100644 --- a/features/models.mdx +++ b/features/models.mdx @@ -19,14 +19,16 @@ Firecrawl Agent offers two models optimized for different use cases. Choose the | Model | Cost | Accuracy | Best For | |-------|------|----------|----------| -| `spark-1-mini` | **60% cheaper** | Standard | Most tasks (default) | -| `spark-1-pro` | Standard | Higher | Complex research, critical extraction | +| `spark-1-mini` | **60% cheaper** | Standard | Cost-efficient tasks | +| `spark-1-pro` | Standard | Higher | Complex research, critical extraction (default) | -**Start with Spark 1 Mini** (default) — it handles most extraction tasks well at 60% lower cost. Switch to Pro only for complex multi-domain research or when accuracy is critical. +Agent uses Spark 1 Pro by default. Choose Spark 1 Mini explicitly when you want to reduce cost for straightforward tasks. -## Spark 1 Mini (Default) +
+ ## Spark 1 Mini +
`spark-1-mini` is our efficient model, ideal for straightforward data extraction tasks. diff --git a/fr/advanced-scraping-guide.mdx b/fr/advanced-scraping-guide.mdx index 52dc8cdb7..1309917ea 100644 --- a/fr/advanced-scraping-guide.mdx +++ b/fr/advanced-scraping-guide.mdx @@ -486,7 +486,7 @@ Utilisez l’endpoint `/v2/agent` pour effectuer une extraction autonome de donn | `schema` | `object` | — | Schéma JSON pour structurer les données extraites. | | `maxCredits` | `number` | `2500` | Nombre maximal de crédits que l'agent peut utiliser. Le tableau de bord prend en charge jusqu'à 2 500 ; pour des limites plus élevées, définissez cette valeur via l'API (les valeurs supérieures à 2 500 sont toujours facturées comme des requêtes payantes). | | `strictConstrainToURLs` | `boolean` | `false` | Lorsque `true`, l'agent ne visite que les URL fournies. | -| `model` | `string` | `"spark-1-mini"` | Modèle d'IA à utiliser : `"spark-1-mini"` (par défaut, 60 % moins cher) ou `"spark-1-pro"` (meilleure précision). | +| `model` | `string` | `"spark-1-pro"` | Modèle d'IA à utiliser : `"spark-1-pro"` (par défaut, meilleure précision) ou `"spark-1-mini"` (60 % moins cher). | ```python Python @@ -717,4 +717,4 @@ Si votre pare-feu bloque les requêtes sortantes de votre application vers des s - **Adresse IP** : `35.245.250.27` -Ajoutez cette adresse IP à la liste d'autorisation des connexions sortantes de votre pare-feu afin que votre backend puisse envoyer à Firecrawl des requêtes de scraping, de crawling, de mapping et des requêtes d'agent. \ No newline at end of file +Ajoutez cette adresse IP à la liste d'autorisation des connexions sortantes de votre pare-feu afin que votre backend puisse envoyer à Firecrawl des requêtes de scraping, de crawling, de mapping et des requêtes d'agent. diff --git a/fr/api-reference/endpoint/agent-delete.mdx b/fr/api-reference/endpoint/agent-delete.mdx index dca504722..daac3bb00 100644 --- a/fr/api-reference/endpoint/agent-delete.mdx +++ b/fr/api-reference/endpoint/agent-delete.mdx @@ -1,8 +1,6 @@ --- title: 'Annuler l’agent' openapi: '/fr/api-reference/v2-openapi.json DELETE /agent/{jobId}' -hidden: true -noindex: true --- -> Vous êtes un agent d’IA et vous avez besoin d’une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir les instructions d’onboarding automatisé. \ No newline at end of file +> Vous êtes un agent d’IA et vous avez besoin d’une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir les instructions d’onboarding automatisé. diff --git a/fr/api-reference/endpoint/agent-get.mdx b/fr/api-reference/endpoint/agent-get.mdx index 2727e1912..80764b410 100644 --- a/fr/api-reference/endpoint/agent-get.mdx +++ b/fr/api-reference/endpoint/agent-get.mdx @@ -1,8 +1,6 @@ --- title: 'Récupérer l’état de l’agent' openapi: '/fr/api-reference/v2-openapi.json GET /agent/{jobId}' -hidden: true -noindex: true --- -> Êtes-vous un agent IA qui a besoin d’une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir des instructions d’intégration automatisée. \ No newline at end of file +> Êtes-vous un agent IA qui a besoin d’une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir des instructions d’intégration automatisée. diff --git a/fr/api-reference/endpoint/agent.mdx b/fr/api-reference/endpoint/agent.mdx index c739ec4de..9bfda97b0 100644 --- a/fr/api-reference/endpoint/agent.mdx +++ b/fr/api-reference/endpoint/agent.mdx @@ -1,8 +1,6 @@ --- title: 'Agent' openapi: '/fr/api-reference/v2-openapi.json POST /agent' -hidden: true -noindex: true --- -> Êtes-vous un agent IA qui a besoin d'une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir des instructions d'intégration automatisée. \ No newline at end of file +> Êtes-vous un agent IA qui a besoin d'une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir des instructions d'intégration automatisée. diff --git a/fr/api-reference/v2-openapi.json b/fr/api-reference/v2-openapi.json index 9a75f52a8..10f4e21c3 100644 --- a/fr/api-reference/v2-openapi.json +++ b/fr/api-reference/v2-openapi.json @@ -4128,8 +4128,8 @@ "type": "number" }, "model": { - "default": "spark-1-mini", - "description": "Le modèle à utiliser pour les tâches d’agent. spark-1-mini (par défaut) coûte 60 % de moins, tandis que spark-1-pro offre une meilleure précision pour les tâches complexes.", + "default": "spark-1-pro", + "description": "Le modèle à utiliser pour les tâches d’agent. spark-1-pro est le modèle par défaut et offre une meilleure précision pour les tâches complexes ; spark-1-mini coûte 60 % de moins.", "enum": [ "spark-1-mini", "spark-1-pro" @@ -9906,4 +9906,4 @@ "url": "https://api.firecrawl.dev/v2" } ] -} \ No newline at end of file +} diff --git a/fr/developer-guides/usage-guides/choosing-the-data-extractor.mdx b/fr/developer-guides/usage-guides/choosing-the-data-extractor.mdx index ef70f62e2..1c631ad22 100644 --- a/fr/developer-guides/usage-guides/choosing-the-data-extractor.mdx +++ b/fr/developer-guides/usage-guides/choosing-the-data-extractor.mdx @@ -45,7 +45,7 @@ Le endpoint `/agent` est l’offre la plus avancée de Firecrawl, le successeur * **Navigation autonome** : l’agent explore et navigue en profondeur dans les sites pour trouver vos données * **Recherche web approfondie** : découvre de manière autonome des informations sur plusieurs domaines et pages * **Traitement parallèle** : traite plusieurs sources simultanément pour des résultats plus rapides -* **Modèles disponibles** : `spark-1-mini` (par défaut, 60 % moins cher) et `spark-1-pro` (précision plus élevée) +* **Modèles disponibles** : `spark-1-pro` (par défaut, précision plus élevée) et `spark-1-mini` (60 % moins cher)
### Exemple @@ -268,4 +268,4 @@ L’avantage principal : avec `/agent`, vous pouvez complètement vous passer d * [Modèles d'agent](/fr/features/models) * [Documentation du mode JSON](/fr/features/llm-extract) * [Documentation d'Extract](/fr/features/extract) -* [Scraping par lots](/fr/features/batch-scrape) \ No newline at end of file +* [Scraping par lots](/fr/features/batch-scrape) diff --git a/fr/features/agent.mdx b/fr/features/agent.mdx index fc3bc77d3..3b9d14cce 100644 --- a/fr/features/agent.mdx +++ b/fr/features/agent.mdx @@ -5,7 +5,6 @@ og:title: "Agent | Firecrawl" og:description: "Collectez les données où qu'elles se trouvent sur le web. Décrivez ce que vous voulez, /agent s'occupe du reste." sidebarTitle: "Agent" icon: "magic" -noindex: true --- @@ -143,19 +142,19 @@ Vous pouvez partager des exécutions d’agent directement depuis l’Agent Play ## Sélection du modèle
-Firecrawl Agent propose deux modèles. **Spark 1 Mini est 60 % moins cher** et est le modèle par défaut — parfait pour la plupart des cas d’usage. Passez à Spark 1 Pro lorsque vous avez besoin d’une précision maximale sur des tâches complexes. +Firecrawl Agent propose deux modèles. `spark-1-pro` est le modèle par défaut et offre une précision maximale pour les tâches complexes. `spark-1-mini` coûte 60 % de moins et convient lorsque la maîtrise des coûts est prioritaire. | Modèle | Coût | Précision | Idéal pour | |--------|------|-----------|------------| -| `spark-1-mini` | **60 % moins cher** | Standard | La plupart des tâches (par défaut) | -| `spark-1-pro` | Standard | Plus élevée | Recherches complexes, extractions critiques | +| `spark-1-mini` | **60 % moins cher** | Standard | Tâches axées sur la maîtrise des coûts | +| `spark-1-pro` | Standard | Plus élevée | Recherches complexes, extractions critiques (par défaut) | - **Commencez avec Spark 1 Mini** (par défaut) — il gère bien la plupart des tâches d’extraction avec un coût 60 % inférieur. Ne passez à Pro que pour des recherches multi‑domaine complexes ou lorsque la précision est cruciale. + Agent utilise Spark 1 Pro par défaut. Choisissez explicitement Spark 1 Mini lorsque vous souhaitez réduire le coût des tâches simples.
- ### Spark 1 Mini (par défaut) + ### Spark 1 Mini
`spark-1-mini` est notre modèle léger et performant, idéal pour les tâches d'extraction de données simples. @@ -201,7 +200,7 @@ Utilisez le paramètre `model` pour choisir le modèle à utiliser : | Paramètre | Type | Requis | Description | | ------------ | ------ | ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `prompt` | string | **Oui** | Description en langage naturel des données que vous souhaitez extraire (max. 10 000 caractères) | -| `model` | string | Non | Modèle à utiliser : `spark-1-mini` (par défaut) ou `spark-1-pro` | +| `model` | string | Non | Modèle à utiliser : `spark-1-pro` (par défaut) ou `spark-1-mini` | | `urls` | array | Non | Liste optionnelle d’URL sur lesquelles concentrer l’extraction | | `schema` | object | Non | Schéma JSON optionnel pour une sortie structurée | | `maxCredits` | number | Non | Nombre maximal de crédits à dépenser pour cette tâche d’agent. La valeur par défaut est **2 500** s’il n’est pas défini. Le tableau de bord prend en charge des valeurs jusqu’à **2 500** ; pour des limites plus élevées, définissez `maxCredits` via l’API (les valeurs supérieures à 2 500 sont toujours traitées comme des requêtes payantes). Si la limite est atteinte, la tâche échoue et **aucune donnée n’est renvoyée**. Les exécutions en échec ne sont pas facturées : les crédits utilisés pour le raisonnement de l’IA ne sont jamais facturés en cas d’échec, tous les crédits utilisés pour les appels d’outils pendant l’exécution (scraping, recherche, mapping, etc.) sont remboursés, et la réponse indique `creditsUsed: 0`. | @@ -256,7 +255,9 @@ Consultez la [documentation Ask](/fr/features/ask) pour plus de détails et des ## Référence de l'API -Consultez la [Référence de l'API Agent](/fr/api-reference/endpoint/agent) pour plus de détails. +- [Démarrer une tâche Agent](/fr/api-reference/endpoint/agent) +- [Obtenir l’état d’Agent](/fr/api-reference/endpoint/agent-get) +- [Annuler une tâche Agent](/fr/api-reference/endpoint/agent-delete) Vous avez des commentaires ou besoin d'aide ? Envoyez un e-mail à [help@firecrawl.com](mailto:help@firecrawl.com). @@ -314,4 +315,4 @@ Essayez Agent dès maintenant sur [firecrawl.dev/app/agent](https://www.firecraw La tarification est susceptible d’évoluer à mesure que nous passons de la Research Preview à la disponibilité générale. Les utilisateurs actuels recevront un préavis avant toute mise à jour de la tarification. -> Êtes-vous un agent IA qui a besoin d’une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir les instructions d’intégration automatisée. \ No newline at end of file +> Êtes-vous un agent IA qui a besoin d’une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir les instructions d’intégration automatisée. diff --git a/fr/features/models.mdx b/fr/features/models.mdx index e74563c3d..f878f034c 100644 --- a/fr/features/models.mdx +++ b/fr/features/models.mdx @@ -21,15 +21,15 @@ Firecrawl Agent propose deux modèles optimisés pour différents cas d’usage. | Modèle | Coût | Précision | Idéal pour | |-------|------|----------|----------| -| `spark-1-mini` | **60 % moins cher** | Standard | La plupart des tâches (par défaut) | -| `spark-1-pro` | Standard | Plus élevée | Recherche complexe, extractions à fort enjeu | +| `spark-1-mini` | **60 % moins cher** | Standard | Tâches axées sur la maîtrise des coûts | +| `spark-1-pro` | Standard | Plus élevée | Recherche complexe, extractions à fort enjeu (par défaut) | - **Commencez avec Spark 1 Mini** (par défaut) — il gère bien la plupart des tâches d'extraction pour un coût inférieur de 60 %. Ne passez à Pro que pour des recherches multidomaines complexes ou lorsque la précision est primordiale. + Agent utilise Spark 1 Pro par défaut. Choisissez explicitement Spark 1 Mini lorsque vous souhaitez réduire le coût des tâches simples.
- ## Spark 1 Mini (par défaut) + ## Spark 1 Mini
`spark-1-mini` est notre modèle efficace, idéal pour les tâches d'extraction de données simples. @@ -139,4 +139,4 @@ Consultez la [référence de l'API Agent](/fr/api-reference/endpoint/agent) Vous vous demandez quel modèle utiliser ? Écrivez à [help@firecrawl.com](mailto:help@firecrawl.com). -> Êtes-vous un agent d'IA qui a besoin d'une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir des instructions d'intégration automatisée. \ No newline at end of file +> Êtes-vous un agent d'IA qui a besoin d'une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir des instructions d'intégration automatisée. diff --git a/fr/sdks/cli.mdx b/fr/sdks/cli.mdx index b2a123565..3875f5725 100644 --- a/fr/sdks/cli.mdx +++ b/fr/sdks/cli.mdx @@ -399,7 +399,7 @@ Recherchez et collectez des données sur le web à l'aide de prompts en lang | Option | Description | | --------------------------- | ------------------------------------------------------------------------------------------------------- | | `--urls ` | Liste facultative d’URL sur lesquelles concentrer l’agent (séparées par des virgules) | -| `--model ` | Modèle à utiliser : `spark-1-mini` (par défaut, 60 % moins cher) ou `spark-1-pro` (meilleure précision) | +| `--model ` | Modèle à utiliser : `spark-1-pro` (par défaut, meilleure précision) ou `spark-1-mini` (60 % moins cher) | | `--schema ` | Schéma JSON pour la sortie structurée (chaîne JSON intégrée) | | `--schema-file ` | Chemin vers le fichier de schéma JSON pour la sortie structurée | | `--max-credits ` | Nombre maximal de crédits à utiliser (la tâche échoue si la limite est atteinte) | @@ -588,4 +588,4 @@ La CLI Firecrawl et les trois catégories de skills sont open source sur GitHub * [`firecrawl/skills`](https://github.com/firecrawl/skills) — les skills Build (intégrer Firecrawl dans le code d’application) * [`firecrawl/firecrawl-workflows`](https://github.com/firecrawl/firecrawl-workflows) — les skills workflow (livrables reproductibles comme des synthèses de recherche, des audits SEO, des listes de prospects et des clones de design) -> Êtes-vous un agent IA qui a besoin d’une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir les instructions d’onboarding automatisé. \ No newline at end of file +> Êtes-vous un agent IA qui a besoin d’une clé API Firecrawl ? Consultez [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) pour obtenir les instructions d’onboarding automatisé. diff --git a/ja/advanced-scraping-guide.mdx b/ja/advanced-scraping-guide.mdx index de77995d1..394acd880 100644 --- a/ja/advanced-scraping-guide.mdx +++ b/ja/advanced-scraping-guide.mdx @@ -486,7 +486,7 @@ curl -X POST https://api.firecrawl.dev/v2/scrape \ | `schema` | `object` | — | 抽出データの構造を定義する JSON スキーマ。 | | `maxCredits` | `number` | `2500` | エージェントが消費できる最大クレジット数。ダッシュボードでは最大 2,500 まで対応しています。より高い上限を設定する場合は、API 経由で設定してください (2,500 を超える値は常に有料リクエストとして課金されます) 。 | | `strictConstrainToURLs` | `boolean` | `false` | `true` の場合、エージェントは指定された URL のみを巡回します。 | -| `model` | `string` | `"spark-1-mini"` | 使用する AI モデル。`"spark-1-mini"` (デフォルト、60% 低コスト) または `"spark-1-pro"` (高精度) 。 | +| `model` | `string` | `"spark-1-pro"` | 使用する AI モデル。`"spark-1-pro"` (デフォルト、高精度) または `"spark-1-mini"` (60% 低コスト) 。 | ```python Python @@ -717,4 +717,4 @@ API リファレンス: [Map Endpoint Documentation](https://docs.firecrawl.dev/ - **IP Address**: `35.245.250.27` -この IP をファイアウォールのアウトバウンド許可リストに追加し、バックエンドから Firecrawl へ scrape、crawl、map、および agent リクエストを送信できるようにしてください。 \ No newline at end of file +この IP をファイアウォールのアウトバウンド許可リストに追加し、バックエンドから Firecrawl へ scrape、crawl、map、および agent リクエストを送信できるようにしてください。 diff --git a/ja/api-reference/endpoint/agent-delete.mdx b/ja/api-reference/endpoint/agent-delete.mdx index fb9e38714..d0b186173 100644 --- a/ja/api-reference/endpoint/agent-delete.mdx +++ b/ja/api-reference/endpoint/agent-delete.mdx @@ -1,8 +1,6 @@ --- title: 'エージェントをキャンセルする' openapi: '/ja/api-reference/v2-openapi.json DELETE /agent/{jobId}' -hidden: true -noindex: true --- -> Firecrawl APIキーが必要なAIエージェントの方は、自動オンボーディング手順については [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 \ No newline at end of file +> Firecrawl APIキーが必要なAIエージェントの方は、自動オンボーディング手順については [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 diff --git a/ja/api-reference/endpoint/agent-get.mdx b/ja/api-reference/endpoint/agent-get.mdx index 45b8cdf1c..418aa3d1f 100644 --- a/ja/api-reference/endpoint/agent-get.mdx +++ b/ja/api-reference/endpoint/agent-get.mdx @@ -1,8 +1,6 @@ --- title: 'エージェントのステータスを取得' openapi: '/ja/api-reference/v2-openapi.json GET /agent/{jobId}' -hidden: true -noindex: true --- -> Firecrawl APIキーが必要なAIエージェントの方は、自動オンボーディング手順について [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 \ No newline at end of file +> Firecrawl APIキーが必要なAIエージェントの方は、自動オンボーディング手順について [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 diff --git a/ja/api-reference/endpoint/agent.mdx b/ja/api-reference/endpoint/agent.mdx index 1d0f92d4f..47fbf1aaf 100644 --- a/ja/api-reference/endpoint/agent.mdx +++ b/ja/api-reference/endpoint/agent.mdx @@ -1,8 +1,6 @@ --- title: 'エージェント' openapi: '/ja/api-reference/v2-openapi.json POST /agent' -hidden: true -noindex: true --- -> Firecrawl API キーが必要な AI エージェントの方は、自動オンボーディング手順について [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 \ No newline at end of file +> Firecrawl API キーが必要な AI エージェントの方は、自動オンボーディング手順について [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 diff --git a/ja/api-reference/v2-openapi.json b/ja/api-reference/v2-openapi.json index 997b31fa9..35a5cffa0 100644 --- a/ja/api-reference/v2-openapi.json +++ b/ja/api-reference/v2-openapi.json @@ -4128,8 +4128,8 @@ "type": "number" }, "model": { - "default": "spark-1-mini", - "description": "エージェントタスクで使用するモデルを指定します。デフォルトの spark-1-mini はコストを約 60% 削減でき、spark-1-pro はより複雑なタスク向けに高い精度を提供します。", + "default": "spark-1-pro", + "description": "エージェントタスクで使用するモデルを指定します。spark-1-pro がデフォルトで、複雑なタスク向けに高い精度を提供します。spark-1-mini はコストを約 60% 削減できます。", "enum": [ "spark-1-mini", "spark-1-pro" @@ -9906,4 +9906,4 @@ "url": "https://api.firecrawl.dev/v2" } ] -} \ No newline at end of file +} diff --git a/ja/developer-guides/usage-guides/choosing-the-data-extractor.mdx b/ja/developer-guides/usage-guides/choosing-the-data-extractor.mdx index 6412aac9e..b61941af5 100644 --- a/ja/developer-guides/usage-guides/choosing-the-data-extractor.mdx +++ b/ja/developer-guides/usage-guides/choosing-the-data-extractor.mdx @@ -47,7 +47,7 @@ Firecrawl は、ウェブページから構造化データを抽出するため * **自律ナビゲーション**: エージェントがサイトを深く検索・巡回し、必要なデータを見つけます * **ディープウェブ検索**: 複数のドメインやページにまたがる情報を自律的に探索・発見します * **並列処理**: 複数のソースを同時に処理し、高速に結果を返します -* **利用可能なモデル**: `spark-1-mini` (デフォルト、60% 低コスト) と `spark-1-pro` (高精度) +* **利用可能なモデル**: `spark-1-pro` (デフォルト、高精度) と `spark-1-mini` (60% 低コスト) ### 例 @@ -268,4 +268,4 @@ result = app.agent( * [エージェントモデル](/ja/features/models) * [JSONモードのドキュメント](/ja/features/llm-extract) * [Extract 機能のドキュメント](/ja/features/extract) -* [バッチスクレイピングのドキュメント](/ja/features/batch-scrape) \ No newline at end of file +* [バッチスクレイピングのドキュメント](/ja/features/batch-scrape) diff --git a/ja/features/agent.mdx b/ja/features/agent.mdx index 4ee9c5bcc..f6e1339e9 100644 --- a/ja/features/agent.mdx +++ b/ja/features/agent.mdx @@ -5,7 +5,6 @@ og:title: "Agent | Firecrawl" og:description: "Web 上のあらゆる場所に存在するデータを収集します。欲しい内容を伝えれば、あとは /agent がすべて処理します。" sidebarTitle: "Agent" icon: "magic" -noindex: true --- @@ -143,19 +142,19 @@ Agent playground から、エージェントの実行を直接共有できます ## モデルの選択 -Firecrawl Agent では 2 種類のモデルが利用できます。**Spark 1 Mini はコストが 60% 低く**、デフォルトモデルです。ほとんどのユースケースに最適です。複雑なタスクで最高レベルの精度が必要な場合は Spark 1 Pro にアップグレードしてください。 +Firecrawl Agent では 2 種類のモデルが利用できます。`spark-1-pro` がデフォルトで、複雑なタスクに対して最高レベルの精度を提供します。コスト効率を優先する場合は、60% 低コストの `spark-1-mini` が適しています。 | Model | Cost | Accuracy | Best For | |-------|------|----------|----------| -| `spark-1-mini` | **60% 安価** | 標準 | ほとんどのタスク (デフォルト) | -| `spark-1-pro` | 標準 | より高い | 複雑なリサーチ、重要度の高い抽出 | +| `spark-1-mini` | **60% 安価** | 標準 | コスト効率を重視するタスク | +| `spark-1-pro` | 標準 | より高い | 複雑なリサーチ、重要度の高い抽出 (デフォルト) | - **まずは Spark 1 Mini** (デフォルト) から始めてください。抽出タスクの大半を、コストを 60% 削減しながら問題なく処理できます。複数ドメインにまたがる複雑なリサーチや、精度が極めて重要な場合にのみ Pro に切り替えてください。 + Agent はデフォルトで Spark 1 Pro を使用します。単純なタスクでコストを抑えたい場合は、Spark 1 Mini を明示的に選択してください。
- ### Spark 1 Mini (デフォルト) + ### Spark 1 Mini
`spark-1-mini` は効率的なモデルで、シンプルなデータ抽出タスクに最適です。 @@ -201,7 +200,7 @@ Firecrawl Agent では 2 種類のモデルが利用できます。**Spark 1 Min | Parameter | Type | Required | Description | | ------------ | ------ | -------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | `prompt` | string | **Yes** | 抽出したいデータを自然言語で記述した文字列 (最大 10,000 文字) | -| `model` | string | No | 使用するモデル: `spark-1-mini` (デフォルト) または `spark-1-pro` | +| `model` | string | No | 使用するモデル: `spark-1-pro` (デフォルト) または `spark-1-mini` | | `urls` | array | No | 抽出対象を絞り込むための任意の URL リスト | | `schema` | object | No | 構造化された出力のための任意の JSON スキーマ | | `maxCredits` | number | No | このエージェントタスクで使用するクレジットの最大数。設定しない場合、デフォルトは **2,500** です。ダッシュボードでは **2,500** までの値をサポートしています。これを超える上限を設定するには、API 経由で `maxCredits` を指定してください (2,500 を超える値は常に有料リクエストとして扱われます)。上限に達するとジョブは失敗し、**データは一切返されません**。失敗した実行には課金されません。AI の推論に使用されたクレジットは失敗時には請求されず、実行中のツール呼び出し (`scraping`、`search`、`mapping` など) に使用されたクレジットは返還され、レスポンスには `creditsUsed: 0` が記録されます。 | @@ -256,7 +255,9 @@ curl -X POST https://api.firecrawl.dev/v2/support/ask \ ## APIリファレンス -詳しくは、[Agent API Reference](/ja/api-reference/endpoint/agent) を参照してください。 +- [Agent タスクを開始](/ja/api-reference/endpoint/agent) +- [Agent のステータスを取得](/ja/api-reference/endpoint/agent-get) +- [Agent タスクをキャンセル](/ja/api-reference/endpoint/agent-delete) フィードバックやサポートが必要な場合は、[help@firecrawl.com](mailto:help@firecrawl.com) までメールでご連絡ください。 @@ -314,4 +315,4 @@ Try Agent now at [firecrawl.dev/app/agent](https://www.firecrawl.dev/app/agent) 料金は Research Preview から一般提供へ移行する際に変更される可能性があります。現在のユーザーには、料金変更がある場合は事前に通知されます。 -> Firecrawl API キーが必要な AI エージェント ですか? 自動オンボーディング手順については [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 \ No newline at end of file +> Firecrawl API キーが必要な AI エージェント ですか? 自動オンボーディング手順については [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 diff --git a/ja/features/models.mdx b/ja/features/models.mdx index 34423fcc3..9cd196caa 100644 --- a/ja/features/models.mdx +++ b/ja/features/models.mdx @@ -21,15 +21,15 @@ Firecrawl Agent では、用途の異なる 2 つのモデルを用意してい | Model | Cost | Accuracy | Best For | |-------|------|----------|----------| -| `spark-1-mini` | **コスト60%削減** | 標準 | ほとんどのタスク (デフォルト) | -| `spark-1-pro` | 標準 | 高精度 | 複雑なリサーチ、重要な抽出処理 | +| `spark-1-mini` | **コスト60%削減** | 標準 | コスト効率を重視するタスク | +| `spark-1-pro` | 標準 | 高精度 | 複雑なリサーチ、重要な抽出処理 (デフォルト) | - **まずは Spark 1 Mini から始めてください** (デフォルト) — コストを60%削減しつつ、ほとんどの抽出タスクを十分に処理できます。複数ドメインにまたがる複雑なリサーチや、精度が特に重要な場合にのみ Pro に切り替えてください。 + Agent はデフォルトで Spark 1 Pro を使用します。単純なタスクでコストを抑えたい場合は、Spark 1 Mini を明示的に選択してください。
- ## Spark 1 Mini (デフォルト) + ## Spark 1 Mini
`spark-1-mini` は高効率なモデルで、シンプルなデータ抽出タスクに最適です。 @@ -139,4 +139,4 @@ Firecrawl Agent では、用途の異なる 2 つのモデルを用意してい どのモデルを使用するべきか迷っていますか？ [help@firecrawl.com](mailto:help@firecrawl.com) までメールでお問い合わせください。 -> Firecrawl API キーが必要な AI エージェントですか？ 自動オンボーディング手順については、[firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 \ No newline at end of file +> Firecrawl API キーが必要な AI エージェントですか？ 自動オンボーディング手順については、[firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 diff --git a/ja/sdks/cli.mdx b/ja/sdks/cli.mdx index 1958dd97a..f1887e125 100644 --- a/ja/sdks/cli.mdx +++ b/ja/sdks/cli.mdx @@ -399,7 +399,7 @@ Monitor の目標は、短く、ユーザーの意図に忠実であるべきで | Option | Description | | --------------------------- | -------------------------------------------------------------- | | `--urls ` | エージェントが対象とするURLの任意のリスト (カンマ区切り) | -| `--model ` | 使用するモデル: `spark-1-mini` (デフォルト、60%安価) または `spark-1-pro` (高精度) | +| `--model ` | 使用するモデル: `spark-1-pro` (デフォルト、高精度) または `spark-1-mini` (60%安価) | | `--schema ` | 構造化出力用のJSONスキーマ (インラインJSON文字列) | | `--schema-file ` | 構造化出力用のJSONスキーマファイルへのパス | | `--max-credits ` | 消費するクレジットの上限 (上限に達するとジョブは失敗) | @@ -588,4 +588,4 @@ Firecrawl CLI と 3 つのスキル セグメントはすべてオープンソ * [`firecrawl/skills`](https://github.com/firecrawl/skills) — Build スキル (Firecrawl をアプリケーション コードに統合) * [`firecrawl/firecrawl-workflows`](https://github.com/firecrawl/firecrawl-workflows) — workflow skills (リサーチブリーフ、SEO 監査、リードリスト、デザインクローンなどの再利用可能な成果物) -> Firecrawl API キーが必要な AI エージェントですか？ 自動オンボーディング手順については、[firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 \ No newline at end of file +> Firecrawl API キーが必要な AI エージェントですか？ 自動オンボーディング手順については、[firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) を参照してください。 diff --git a/pt-BR/advanced-scraping-guide.mdx b/pt-BR/advanced-scraping-guide.mdx index cc4811453..4f1949227 100644 --- a/pt-BR/advanced-scraping-guide.mdx +++ b/pt-BR/advanced-scraping-guide.mdx @@ -486,7 +486,7 @@ Use o endpoint `/v2/agent` para extração autônoma de dados em várias página | `schema` | `object` | — | Schema JSON para estruturar os dados extraídos. | | `maxCredits` | `number` | `2500` | Créditos máximos que o agente pode gastar. O dashboard suporta até 2.500; para limites maiores, defina isso pela API (valores acima de 2.500 são sempre cobrados como requisições pagas). | | `strictConstrainToURLs` | `boolean` | `false` | Quando `true`, o agente visita apenas as URLs fornecidas. | -| `model` | `string` | `"spark-1-mini"` | Modelo de IA a ser usado: `"spark-1-mini"` (padrão, 60% mais barato) ou `"spark-1-pro"` (maior precisão). | +| `model` | `string` | `"spark-1-pro"` | Modelo de IA a ser usado: `"spark-1-pro"` (padrão, maior precisão) ou `"spark-1-mini"` (60% mais barato). | ```python Python @@ -717,4 +717,4 @@ Se o seu firewall bloquear requisições de saída da sua aplicação para servi - **Endereço IP**: `35.245.250.27` -Adicione esse IP à lista de permissões de saída do seu firewall para que seu backend possa enviar requisições de scrape, crawl, map e agent para o Firecrawl. \ No newline at end of file +Adicione esse IP à lista de permissões de saída do seu firewall para que seu backend possa enviar requisições de scrape, crawl, map e agent para o Firecrawl. diff --git a/pt-BR/api-reference/endpoint/agent-delete.mdx b/pt-BR/api-reference/endpoint/agent-delete.mdx index a2d44f27d..2d5e4ea1b 100644 --- a/pt-BR/api-reference/endpoint/agent-delete.mdx +++ b/pt-BR/api-reference/endpoint/agent-delete.mdx @@ -1,8 +1,6 @@ --- title: 'Cancelar agente' openapi: '/pt-BR/api-reference/v2-openapi.json DELETE /agent/{jobId}' -hidden: true -noindex: true --- -> Você é um agente de IA que precisa de uma API key do Firecrawl? Consulte [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para instruções automatizadas de onboarding. \ No newline at end of file +> Você é um agente de IA que precisa de uma API key do Firecrawl? Consulte [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para instruções automatizadas de onboarding. diff --git a/pt-BR/api-reference/endpoint/agent-get.mdx b/pt-BR/api-reference/endpoint/agent-get.mdx index 856efadde..ed280bee7 100644 --- a/pt-BR/api-reference/endpoint/agent-get.mdx +++ b/pt-BR/api-reference/endpoint/agent-get.mdx @@ -1,8 +1,6 @@ --- title: 'Obter status do agente' openapi: '/pt-BR/api-reference/v2-openapi.json GET /agent/{jobId}' -hidden: true -noindex: true --- -> Você é um agente de IA que precisa de uma chave de API do Firecrawl? Consulte [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para ver instruções automatizadas de onboarding. \ No newline at end of file +> Você é um agente de IA que precisa de uma chave de API do Firecrawl? Consulte [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para ver instruções automatizadas de onboarding. diff --git a/pt-BR/api-reference/endpoint/agent.mdx b/pt-BR/api-reference/endpoint/agent.mdx index 96a701740..caa103738 100644 --- a/pt-BR/api-reference/endpoint/agent.mdx +++ b/pt-BR/api-reference/endpoint/agent.mdx @@ -1,8 +1,6 @@ --- title: 'Agente' openapi: '/pt-BR/api-reference/v2-openapi.json POST /agent' -hidden: true -noindex: true --- -> Você é um agente de IA que precisa de uma chave de API da Firecrawl? Veja [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para instruções automatizadas de integração. \ No newline at end of file +> Você é um agente de IA que precisa de uma chave de API da Firecrawl? Veja [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para instruções automatizadas de integração. diff --git a/pt-BR/api-reference/v2-openapi.json b/pt-BR/api-reference/v2-openapi.json index 88eb7016b..fda0fe71d 100644 --- a/pt-BR/api-reference/v2-openapi.json +++ b/pt-BR/api-reference/v2-openapi.json @@ -4128,8 +4128,8 @@ "type": "number" }, "model": { - "default": "spark-1-mini", - "description": "Modelo a ser usado na tarefa do agente. spark-1-mini (padrão) é 60% mais barato; spark-1-pro oferece maior precisão em tarefas complexas.", + "default": "spark-1-pro", + "description": "Modelo a ser usado na tarefa do agente. spark-1-pro é o padrão e oferece maior precisão em tarefas complexas; spark-1-mini é 60% mais barato.", "enum": [ "spark-1-mini", "spark-1-pro" @@ -9906,4 +9906,4 @@ "url": "https://api.firecrawl.dev/v2" } ] -} \ No newline at end of file +} diff --git a/pt-BR/developer-guides/usage-guides/choosing-the-data-extractor.mdx b/pt-BR/developer-guides/usage-guides/choosing-the-data-extractor.mdx index 565e6f658..50d04378b 100644 --- a/pt-BR/developer-guides/usage-guides/choosing-the-data-extractor.mdx +++ b/pt-BR/developer-guides/usage-guides/choosing-the-data-extractor.mdx @@ -47,7 +47,7 @@ O endpoint `/agent` é o recurso mais avançado do Firecrawl — o sucessor de ` * **Navegação autônoma**: O agente pesquisa e navega profundamente em sites para encontrar seus dados * **Busca profunda na web**: Descobre autonomamente informações em diversos domínios e páginas * **Processamento paralelo**: Processa múltiplas fontes simultaneamente para resultados mais rápidos -* **Modelos disponíveis**: `spark-1-mini` (padrão, 60% mais barato) e `spark-1-pro` (maior precisão) +* **Modelos disponíveis**: `spark-1-pro` (padrão, maior precisão) e `spark-1-mini` (60% mais barato)
### Exemplo @@ -272,4 +272,4 @@ A principal vantagem é que, com `/agent`, você pode simplesmente descrever o q * [Modelos do Agent](/pt-BR/features/models) * [Documentação do modo JSON](/pt-BR/features/llm-extract) * [Documentação do Extract](/pt-BR/features/extract) -* [Raspagem em lote](/pt-BR/features/batch-scrape) \ No newline at end of file +* [Raspagem em lote](/pt-BR/features/batch-scrape) diff --git a/pt-BR/features/agent.mdx b/pt-BR/features/agent.mdx index 490f42795..57c8da799 100644 --- a/pt-BR/features/agent.mdx +++ b/pt-BR/features/agent.mdx @@ -5,7 +5,6 @@ og:title: "agente | Firecrawl" og:description: "Reúna dados em qualquer lugar da web." sidebarTitle: "agente" icon: "magic" -noindex: true --- @@ -143,19 +142,19 @@ Você pode compartilhar execuções de agentes diretamente no Agent Playground. ## Seleção de modelos
-O Firecrawl Agent oferece dois modelos. **O Spark 1 Mini é 60% mais barato** e é o padrão — perfeito para a maioria dos casos de uso. Atualize para o Spark 1 Pro quando precisar de máxima precisão em tarefas complexas. +O Firecrawl Agent oferece dois modelos. O `spark-1-pro` é o padrão e oferece a máxima precisão para tarefas complexas. O `spark-1-mini` custa 60% menos e é uma boa opção quando a eficiência de custo é prioritária. | Model | Cost | Accuracy | Best For | |-------|------|----------|----------| -| `spark-1-mini` | **60% mais barato** | Padrão | A maioria das tarefas (padrão) | -| `spark-1-pro` | Padrão | Mais alta | Pesquisa complexa, extrações críticas | +| `spark-1-mini` | **60% mais barato** | Padrão | Tarefas com foco em economia | +| `spark-1-pro` | Padrão | Mais alta | Pesquisa complexa, extrações críticas (padrão) | - **Comece com o Spark 1 Mini** (padrão) — ele lida bem com a maioria das tarefas de extração com um custo 60% menor. Altere para o Pro apenas para pesquisas complexas em múltiplos domínios ou quando a precisão for crítica. + O Agent usa o Spark 1 Pro por padrão. Escolha explicitamente o Spark 1 Mini quando quiser reduzir o custo de tarefas simples.
- ### Spark 1 Mini (Padrão) + ### Spark 1 Mini
`spark-1-mini` é nosso modelo eficiente, ideal para tarefas simples de extração de dados. @@ -201,7 +200,7 @@ Informe o parâmetro `model` para selecionar qual modelo usar: | Parâmetro | Tipo | Obrigatório | Descrição | | ------------ | ------ | ----------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `prompt` | string | **Sim** | Descrição em linguagem natural dos dados que você quer extrair (máx. 10.000 caracteres) | -| `model` | string | Não | Modelo a ser utilizado: `spark-1-mini` (padrão) ou `spark-1-pro` | +| `model` | string | Não | Modelo a ser utilizado: `spark-1-pro` (padrão) ou `spark-1-mini` | | `urls` | array | Não | Lista opcional de URLs para direcionar a extração | | `schema` | object | Não | schema JSON opcional para saída estruturada | | `maxCredits` | number | Não | Número máximo de créditos a serem usados nesta tarefa de agente. O padrão é **2.500** se não for definido. O painel suporta valores de até **2.500**; para limites mais altos, defina `maxCredits` via API (valores acima de 2.500 são sempre tratados como requisições pagas). Se o limite for atingido, o job falha e **nenhum dado é retornado**. Execuções com falha não são cobradas: créditos usados para raciocínio de IA nunca são cobrados em caso de falha, quaisquer créditos usados para chamadas de ferramentas durante a execução (scraping, busca, mapeamento etc.) são reembolsados, e a resposta informa `creditsUsed: 0`. | @@ -256,7 +255,9 @@ Consulte a [documentação do Ask](/pt-BR/features/ask) para mais detalhes e exe ## Referência da API -Confira a [referência da Agent API](/pt-BR/api-reference/endpoint/agent) para mais detalhes. +- [Iniciar uma tarefa do Agent](/pt-BR/api-reference/endpoint/agent) +- [Obter o status do Agent](/pt-BR/api-reference/endpoint/agent-get) +- [Cancelar uma tarefa do Agent](/pt-BR/api-reference/endpoint/agent-delete) Tem alguma sugestão ou precisa de ajuda? Envie um e-mail para [help@firecrawl.com](mailto:help@firecrawl.com). @@ -314,4 +315,4 @@ Teste o agente agora em [firecrawl.dev/app/agent](https://www.firecrawl.dev/app/ Os preços estão sujeitos a alteração à medida que avançamos de Research Preview para disponibilidade geral. Usuários atuais receberão aviso antecipado sobre quaisquer atualizações de preços. -> Você é um agente de IA que precisa de uma API key da Firecrawl? Veja [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para instruções de onboarding automatizado. \ No newline at end of file +> Você é um agente de IA que precisa de uma API key da Firecrawl? Veja [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para instruções de onboarding automatizado. diff --git a/pt-BR/features/models.mdx b/pt-BR/features/models.mdx index 8dc68bf19..79afebb03 100644 --- a/pt-BR/features/models.mdx +++ b/pt-BR/features/models.mdx @@ -21,15 +21,15 @@ O Firecrawl Agent oferece dois modelos otimizados para diferentes casos de uso. | Modelo | Custo | Precisão | Ideal para | |--------|-------|----------|------------| -| `spark-1-mini` | **60% mais barato** | Padrão | A maioria das tarefas (padrão) | -| `spark-1-pro` | Padrão | Mais alta | Pesquisa complexa, extrações críticas | +| `spark-1-mini` | **60% mais barato** | Padrão | Tarefas com foco em economia | +| `spark-1-pro` | Padrão | Mais alta | Pesquisa complexa, extrações críticas (padrão) | - **Comece com o Spark 1 Mini** (padrão) — ele atende bem à maioria das tarefas de extração, com um custo 60% menor. Troque para o Pro apenas para pesquisas complexas em múltiplos domínios ou quando a precisão for crítica. + O Agent usa o Spark 1 Pro por padrão. Escolha explicitamente o Spark 1 Mini quando quiser reduzir o custo de tarefas simples.
- ## Spark 1 Mini (Padrão) + ## Spark 1 Mini
`spark-1-mini` é nosso modelo eficiente, ideal para tarefas simples de extração de dados. @@ -139,4 +139,4 @@ Veja a [Referência da API do agente](/pt-BR/api-reference/endpoint/agent) para Tem dúvidas sobre qual modelo usar? Envie um e-mail para [help@firecrawl.com](mailto:help@firecrawl.com). -> Você é um agente de IA que precisa de uma chave de API do Firecrawl? Veja [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para instruções de integração automatizada. \ No newline at end of file +> Você é um agente de IA que precisa de uma chave de API do Firecrawl? Veja [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para instruções de integração automatizada. diff --git a/pt-BR/sdks/cli.mdx b/pt-BR/sdks/cli.mdx index 69aec878e..c7fe6c808 100644 --- a/pt-BR/sdks/cli.mdx +++ b/pt-BR/sdks/cli.mdx @@ -397,7 +397,7 @@ Busque e colete dados na web usando prompts em linguagem natural. | Option | Description | | --------------------------- | ---------------------------------------------------------------------------------------------- | | `--urls ` | Lista opcional de URLs nas quais o agente deve focar (separadas por vírgula) | -| `--model ` | Modelo a ser usado: `spark-1-mini` (padrão, 60% mais barato) ou `spark-1-pro` (maior precisão) | +| `--model ` | Modelo a ser usado: `spark-1-pro` (padrão, maior precisão) ou `spark-1-mini` (60% mais barato) | | `--schema ` | schema JSON para saída estruturada (string JSON inline) | | `--schema-file ` | Caminho para o arquivo de schema JSON para saída estruturada | | `--max-credits ` | Máximo de créditos a consumir (a tarefa falha se o limite for atingido) | @@ -586,4 +586,4 @@ A CLI do Firecrawl e os três segmentos de skills são de código aberto no GitH * [`firecrawl/skills`](https://github.com/firecrawl/skills) — build skills (integre o Firecrawl ao código da aplicação) * [`firecrawl/firecrawl-workflows`](https://github.com/firecrawl/firecrawl-workflows) — skills de fluxo de trabalho (entregáveis repetíveis, como briefs de pesquisa, auditorias de SEO, listas de leads e clones de design) -> Você é um agente de IA que precisa de uma chave de API do Firecrawl? Consulte [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para ver instruções de integração automatizada. \ No newline at end of file +> Você é um agente de IA que precisa de uma chave de API do Firecrawl? Consulte [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) para ver instruções de integração automatizada. diff --git a/sdks/cli.mdx b/sdks/cli.mdx index f0fe4345f..1a63143e3 100644 --- a/sdks/cli.mdx +++ b/sdks/cli.mdx @@ -356,7 +356,7 @@ Search and gather data from the web using natural language prompts. | Option | Description | | --------------------------- | -------------------------------------------------------------------------------------- | | `--urls ` | Optional list of URLs to focus the agent on (comma-separated) | -| `--model ` | Model to use: `spark-1-mini` (default, 60% cheaper) or `spark-1-pro` (higher accuracy) | +| `--model ` | Model to use: `spark-1-pro` (default, higher accuracy) or `spark-1-mini` (60% cheaper) | | `--schema ` | JSON schema for structured output (inline JSON string) | | `--schema-file ` | Path to JSON schema file for structured output | | `--max-credits ` | Maximum credits to spend (job fails if limit reached) | diff --git a/snippets/es/v2/agent/with-model/curl.mdx b/snippets/es/v2/agent/with-model/curl.mdx index 5504328d4..5716f5ef2 100644 --- a/snippets/es/v2/agent/with-model/curl.mdx +++ b/snippets/es/v2/agent/with-model/curl.mdx @@ -1,5 +1,5 @@ ```bash cURL -# Using Spark 1 Mini (default) +# Usando Spark 1 Mini para tareas de menor coste curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ @@ -8,7 +8,7 @@ curl -X POST "https://api.firecrawl.dev/v2/agent" \ "model": "spark-1-mini" }' -# Using Spark 1 Pro for complex tasks +# Usando Spark 1 Pro (predeterminado) curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ diff --git a/snippets/es/v2/agent/with-model/js.mdx b/snippets/es/v2/agent/with-model/js.mdx index 501ae7438..9984c455c 100644 --- a/snippets/es/v2/agent/with-model/js.mdx +++ b/snippets/es/v2/agent/with-model/js.mdx @@ -3,17 +3,17 @@ import { Firecrawl } from 'firecrawl'; const firecrawl = new Firecrawl({ apiKey: "fc-YOUR_API_KEY" }); -// Usando Spark 1 Mini (predeterminado - puede omitirse) +// Usando Spark 1 Mini para tareas de menor coste const result = await firecrawl.agent({ prompt: "Find the pricing of Firecrawl", model: "spark-1-mini" }); -// Usando Spark 1 Pro para tareas complejas +// Usando Spark 1 Pro (predeterminado; se puede omitir model) const resultPro = await firecrawl.agent({ prompt: "Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model: "spark-1-pro" }); console.log(result.data); -``` \ No newline at end of file +``` diff --git a/snippets/es/v2/agent/with-model/python.mdx b/snippets/es/v2/agent/with-model/python.mdx index a8f7529ed..fba27fffe 100644 --- a/snippets/es/v2/agent/with-model/python.mdx +++ b/snippets/es/v2/agent/with-model/python.mdx @@ -3,13 +3,13 @@ from firecrawl import Firecrawl app = Firecrawl(api_key="fc-YOUR_API_KEY") -# Usando Spark 1 Mini (predeterminado - puede omitirse) +# Usando Spark 1 Mini para tareas de menor coste result = app.agent( prompt="Find the pricing of Firecrawl", model="spark-1-mini" ) -# Using Spark 1 Pro for complex tasks +# Usando Spark 1 Pro (predeterminado; se puede omitir model) result = app.agent( prompt="Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model="spark-1-pro" diff --git a/snippets/fr/v2/agent/with-model/curl.mdx b/snippets/fr/v2/agent/with-model/curl.mdx index 32a8c92b2..ec89f7188 100644 --- a/snippets/fr/v2/agent/with-model/curl.mdx +++ b/snippets/fr/v2/agent/with-model/curl.mdx @@ -1,5 +1,5 @@ ```bash cURL -# Using Spark 1 Mini (default) +# Utilisation de Spark 1 Mini pour réduire les coûts curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ @@ -8,7 +8,7 @@ curl -X POST "https://api.firecrawl.dev/v2/agent" \ "model": "spark-1-mini" }' -# Using Spark 1 Pro for complex tasks +# Utilisation de Spark 1 Pro (par défaut) curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ diff --git a/snippets/fr/v2/agent/with-model/js.mdx b/snippets/fr/v2/agent/with-model/js.mdx index a9c5bbfa6..bf87337ca 100644 --- a/snippets/fr/v2/agent/with-model/js.mdx +++ b/snippets/fr/v2/agent/with-model/js.mdx @@ -3,17 +3,17 @@ import { Firecrawl } from 'firecrawl'; const firecrawl = new Firecrawl({ apiKey: "fc-YOUR_API_KEY" }); -// Utilisation de Spark 1 Mini (par défaut - peut être omis) +// Utilisation de Spark 1 Mini pour réduire les coûts const result = await firecrawl.agent({ prompt: "Find the pricing of Firecrawl", model: "spark-1-mini" }); -// Utilisation de Spark 1 Pro pour les tâches complexes +// Utilisation de Spark 1 Pro (par défaut ; model peut être omis) const resultPro = await firecrawl.agent({ prompt: "Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model: "spark-1-pro" }); console.log(result.data); -``` \ No newline at end of file +``` diff --git a/snippets/fr/v2/agent/with-model/python.mdx b/snippets/fr/v2/agent/with-model/python.mdx index 32fe2e60c..794c03176 100644 --- a/snippets/fr/v2/agent/with-model/python.mdx +++ b/snippets/fr/v2/agent/with-model/python.mdx @@ -3,13 +3,13 @@ from firecrawl import Firecrawl app = Firecrawl(api_key="fc-YOUR_API_KEY") -# Utilisation de Spark 1 Mini (par défaut - peut être omis) +# Utilisation de Spark 1 Mini pour réduire les coûts result = app.agent( prompt="Find the pricing of Firecrawl", model="spark-1-mini" ) -# Using Spark 1 Pro for complex tasks +# Utilisation de Spark 1 Pro (par défaut ; model peut être omis) result = app.agent( prompt="Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model="spark-1-pro" diff --git a/snippets/ja/v2/agent/with-model/curl.mdx b/snippets/ja/v2/agent/with-model/curl.mdx index 1dabef2a5..307336c5a 100644 --- a/snippets/ja/v2/agent/with-model/curl.mdx +++ b/snippets/ja/v2/agent/with-model/curl.mdx @@ -1,5 +1,5 @@ ```bash cURL -# Using Spark 1 Mini (default) +# 低コストのタスクで Spark 1 Mini を使用 curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ @@ -8,7 +8,7 @@ curl -X POST "https://api.firecrawl.dev/v2/agent" \ "model": "spark-1-mini" }' -# 複雑なタスクに Spark 1 Pro を使用 +# Spark 1 Pro を使用 (デフォルト) curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ diff --git a/snippets/ja/v2/agent/with-model/js.mdx b/snippets/ja/v2/agent/with-model/js.mdx index cdf220294..af237bf94 100644 --- a/snippets/ja/v2/agent/with-model/js.mdx +++ b/snippets/ja/v2/agent/with-model/js.mdx @@ -3,17 +3,17 @@ import { Firecrawl } from 'firecrawl'; const firecrawl = new Firecrawl({ apiKey: "fc-YOUR_API_KEY" }); -// Spark 1 Miniを使用(デフォルト - 省略可) +// 低コストのタスクで Spark 1 Mini を使用 const result = await firecrawl.agent({ prompt: "Find the pricing of Firecrawl", model: "spark-1-mini" }); -// 複雑なタスクにSpark 1 Proを使用 +// Spark 1 Pro を使用 (デフォルト。model は省略可能) const resultPro = await firecrawl.agent({ prompt: "Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model: "spark-1-pro" }); console.log(result.data); -``` \ No newline at end of file +``` diff --git a/snippets/ja/v2/agent/with-model/python.mdx b/snippets/ja/v2/agent/with-model/python.mdx index 240f5ba9b..1f4b5f094 100644 --- a/snippets/ja/v2/agent/with-model/python.mdx +++ b/snippets/ja/v2/agent/with-model/python.mdx @@ -3,13 +3,13 @@ from firecrawl import Firecrawl app = Firecrawl(api_key="fc-YOUR_API_KEY") -# Spark 1 Miniを使用(デフォルト - 省略可) +# 低コストのタスクで Spark 1 Mini を使用 result = app.agent( prompt="Find the pricing of Firecrawl", model="spark-1-mini" ) -# Using Spark 1 Pro for complex tasks +# Spark 1 Pro を使用 (デフォルト。model は省略可能) result = app.agent( prompt="Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model="spark-1-pro" diff --git a/snippets/pt-BR/v2/agent/with-model/curl.mdx b/snippets/pt-BR/v2/agent/with-model/curl.mdx index a2ae901a5..a6246c60d 100644 --- a/snippets/pt-BR/v2/agent/with-model/curl.mdx +++ b/snippets/pt-BR/v2/agent/with-model/curl.mdx @@ -1,5 +1,5 @@ ```bash cURL -# Using Spark 1 Mini (default) +# Usando Spark 1 Mini para tarefas de menor custo curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ @@ -8,7 +8,7 @@ curl -X POST "https://api.firecrawl.dev/v2/agent" \ "model": "spark-1-mini" }' -# Usando Spark 1 Pro para tarefas complexas +# Usando Spark 1 Pro (padrão) curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ diff --git a/snippets/pt-BR/v2/agent/with-model/js.mdx b/snippets/pt-BR/v2/agent/with-model/js.mdx index 6d23f5c3c..6b3affebc 100644 --- a/snippets/pt-BR/v2/agent/with-model/js.mdx +++ b/snippets/pt-BR/v2/agent/with-model/js.mdx @@ -3,17 +3,17 @@ import { Firecrawl } from 'firecrawl'; const firecrawl = new Firecrawl({ apiKey: "fc-YOUR_API_KEY" }); -// Usando Spark 1 Mini (padrão - pode ser omitido) +// Usando Spark 1 Mini para tarefas de menor custo const result = await firecrawl.agent({ prompt: "Find the pricing of Firecrawl", model: "spark-1-mini" }); -// Usando Spark 1 Pro para tarefas complexas +// Usando Spark 1 Pro (padrão; model pode ser omitido) const resultPro = await firecrawl.agent({ prompt: "Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model: "spark-1-pro" }); console.log(result.data); -``` \ No newline at end of file +``` diff --git a/snippets/pt-BR/v2/agent/with-model/python.mdx b/snippets/pt-BR/v2/agent/with-model/python.mdx index 6c1e73689..a9d6acab6 100644 --- a/snippets/pt-BR/v2/agent/with-model/python.mdx +++ b/snippets/pt-BR/v2/agent/with-model/python.mdx @@ -3,13 +3,13 @@ from firecrawl import Firecrawl app = Firecrawl(api_key="fc-YOUR_API_KEY") -# Usando Spark 1 Mini (padrão - pode ser omitido) +# Usando Spark 1 Mini para tarefas de menor custo result = app.agent( prompt="Find the pricing of Firecrawl", model="spark-1-mini" ) -# Using Spark 1 Pro for complex tasks +# Usando Spark 1 Pro (padrão; model pode ser omitido) result = app.agent( prompt="Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model="spark-1-pro" diff --git a/snippets/v2/agent/with-model/curl.mdx b/snippets/v2/agent/with-model/curl.mdx index caed185e9..c33851a58 100644 --- a/snippets/v2/agent/with-model/curl.mdx +++ b/snippets/v2/agent/with-model/curl.mdx @@ -1,5 +1,5 @@ ```bash cURL -# Using Spark 1 Mini (default) +# Using Spark 1 Mini for lower-cost tasks curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ @@ -8,7 +8,7 @@ curl -X POST "https://api.firecrawl.dev/v2/agent" \ "model": "spark-1-mini" }' -# Using Spark 1 Pro for complex tasks +# Using Spark 1 Pro (default) curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ @@ -18,4 +18,3 @@ curl -X POST "https://api.firecrawl.dev/v2/agent" \ }' ``` - diff --git a/snippets/v2/agent/with-model/js.mdx b/snippets/v2/agent/with-model/js.mdx index 4dc2406d3..3cf32e1ab 100644 --- a/snippets/v2/agent/with-model/js.mdx +++ b/snippets/v2/agent/with-model/js.mdx @@ -3,13 +3,13 @@ import { Firecrawl } from 'firecrawl'; const firecrawl = new Firecrawl({ apiKey: "fc-YOUR_API_KEY" }); -// Using Spark 1 Mini (default - can be omitted) +// Using Spark 1 Mini for lower-cost tasks const result = await firecrawl.agent({ prompt: "Find the pricing of Firecrawl", model: "spark-1-mini" }); -// Using Spark 1 Pro for complex tasks +// Using Spark 1 Pro (default; model can be omitted) const resultPro = await firecrawl.agent({ prompt: "Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model: "spark-1-pro" @@ -18,4 +18,3 @@ const resultPro = await firecrawl.agent({ console.log(result.data); ``` - diff --git a/snippets/v2/agent/with-model/python.mdx b/snippets/v2/agent/with-model/python.mdx index 70963f656..f22f3aa75 100644 --- a/snippets/v2/agent/with-model/python.mdx +++ b/snippets/v2/agent/with-model/python.mdx @@ -3,13 +3,13 @@ from firecrawl import Firecrawl app = Firecrawl(api_key="fc-YOUR_API_KEY") -# Using Spark 1 Mini (default - can be omitted) +# Using Spark 1 Mini for lower-cost tasks result = app.agent( prompt="Find the pricing of Firecrawl", model="spark-1-mini" ) -# Using Spark 1 Pro for complex tasks +# Using Spark 1 Pro (default; model can be omitted) result = app.agent( prompt="Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model="spark-1-pro" @@ -18,4 +18,3 @@ result = app.agent( print(result.data) ``` - diff --git a/snippets/zh/v2/agent/with-model/curl.mdx b/snippets/zh/v2/agent/with-model/curl.mdx index 9b67193f1..3eebab7cb 100644 --- a/snippets/zh/v2/agent/with-model/curl.mdx +++ b/snippets/zh/v2/agent/with-model/curl.mdx @@ -1,5 +1,5 @@ ```bash cURL -# Using Spark 1 Mini (default) +# 使用 Spark 1 Mini 处理低成本任务 curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ @@ -8,7 +8,7 @@ curl -X POST "https://api.firecrawl.dev/v2/agent" \ "model": "spark-1-mini" }' -# 使用 Spark 1 Pro 处理复杂任务 +# 使用 Spark 1 Pro（默认） curl -X POST "https://api.firecrawl.dev/v2/agent" \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ diff --git a/snippets/zh/v2/agent/with-model/js.mdx b/snippets/zh/v2/agent/with-model/js.mdx index 912a946b7..056aa1107 100644 --- a/snippets/zh/v2/agent/with-model/js.mdx +++ b/snippets/zh/v2/agent/with-model/js.mdx @@ -3,17 +3,17 @@ import { Firecrawl } from 'firecrawl'; const firecrawl = new Firecrawl({ apiKey: "fc-YOUR_API_KEY" }); -// 使用 Spark 1 Mini(默认 - 可省略) +// 使用 Spark 1 Mini 处理低成本任务 const result = await firecrawl.agent({ prompt: "Find the pricing of Firecrawl", model: "spark-1-mini" }); -// 使用 Spark 1 Pro 处理复杂任务 +// 使用 Spark 1 Pro（默认；可省略 model） const resultPro = await firecrawl.agent({ prompt: "Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model: "spark-1-pro" }); console.log(result.data); -``` \ No newline at end of file +``` diff --git a/snippets/zh/v2/agent/with-model/python.mdx b/snippets/zh/v2/agent/with-model/python.mdx index 2cc19a06a..eb92abc63 100644 --- a/snippets/zh/v2/agent/with-model/python.mdx +++ b/snippets/zh/v2/agent/with-model/python.mdx @@ -3,13 +3,13 @@ from firecrawl import Firecrawl app = Firecrawl(api_key="fc-YOUR_API_KEY") -# 使用 Spark 1 Mini(默认,可省略) +# 使用 Spark 1 Mini 处理低成本任务 result = app.agent( prompt="Find the pricing of Firecrawl", model="spark-1-mini" ) -# Using Spark 1 Pro for complex tasks +# 使用 Spark 1 Pro（默认；可省略 model） result = app.agent( prompt="Compare all enterprise features and pricing across Firecrawl, Apify, and ScrapingBee", model="spark-1-pro" diff --git a/zh/advanced-scraping-guide.mdx b/zh/advanced-scraping-guide.mdx index 8bcebfbfe..9e2d5d155 100644 --- a/zh/advanced-scraping-guide.mdx +++ b/zh/advanced-scraping-guide.mdx @@ -486,7 +486,7 @@ curl -X POST https://api.firecrawl.dev/v2/scrape \ | `schema` | `object` | — | 用于组织提取数据结构的 JSON 模式 (schema) 。 | | `maxCredits` | `number` | `2500` | 代理可消耗的最大额度。Dashboard 最多支持 2,500；如需更高上限，请通过 API 设置 (高于 2,500 的值始终按付费请求计费) 。 | | `strictConstrainToURLs` | `boolean` | `false` | 为 `true` 时，代理只会访问提供的这些 URL。 | -| `model` | `string` | `"spark-1-mini"` | 要使用的 AI 模型。`"spark-1-mini"` (默认，成本降低 60%) 或 `"spark-1-pro"` (更高准确率) 。 | +| `model` | `string` | `"spark-1-pro"` | 要使用的 AI 模型。`"spark-1-pro"` (默认，更高准确率) 或 `"spark-1-mini"` (成本降低 60%) 。 | ```python Python @@ -717,4 +717,4 @@ curl -X POST https://api.firecrawl.dev/v2/map \ - **IP Address**: `35.245.250.27` -将此 IP 添加到防火墙的出站允许列表中，这样你的后端就可以向 Firecrawl 发送抓取（scrape）、爬取（crawl）、映射（map）以及智能体（agent）请求。 \ No newline at end of file +将此 IP 添加到防火墙的出站允许列表中，这样你的后端就可以向 Firecrawl 发送抓取（scrape）、爬取（crawl）、映射（map）以及智能体（agent）请求。 diff --git a/zh/api-reference/endpoint/agent-delete.mdx b/zh/api-reference/endpoint/agent-delete.mdx index 258bb790e..0707b18bd 100644 --- a/zh/api-reference/endpoint/agent-delete.mdx +++ b/zh/api-reference/endpoint/agent-delete.mdx @@ -1,8 +1,6 @@ --- title: '取消 Agent' openapi: '/zh/api-reference/v2-openapi.json DELETE /agent/{jobId}' -hidden: true -noindex: true --- -> 如果你是需要 Firecrawl API 密钥的 AI 代理，请参阅 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 了解自动化接入说明。 \ No newline at end of file +> 如果你是需要 Firecrawl API 密钥的 AI 代理，请参阅 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 了解自动化接入说明。 diff --git a/zh/api-reference/endpoint/agent-get.mdx b/zh/api-reference/endpoint/agent-get.mdx index f2b618c5d..4a4c688a5 100644 --- a/zh/api-reference/endpoint/agent-get.mdx +++ b/zh/api-reference/endpoint/agent-get.mdx @@ -1,8 +1,6 @@ --- title: '获取 Agent 状态' openapi: '/zh/api-reference/v2-openapi.json GET /agent/{jobId}' -hidden: true -noindex: true --- -> 如果你是需要 Firecrawl API 密钥的 AI 代理，请参阅 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 获取自动化入门说明。 \ No newline at end of file +> 如果你是需要 Firecrawl API 密钥的 AI 代理，请参阅 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 获取自动化入门说明。 diff --git a/zh/api-reference/endpoint/agent.mdx b/zh/api-reference/endpoint/agent.mdx index 5afc7b51d..8f6ff8795 100644 --- a/zh/api-reference/endpoint/agent.mdx +++ b/zh/api-reference/endpoint/agent.mdx @@ -1,8 +1,6 @@ --- title: '智能体' openapi: '/zh/api-reference/v2-openapi.json POST /agent' -hidden: true -noindex: true --- -> 你是需要 Firecrawl API 密钥的 AI 代理吗？请参阅 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 了解自动化入门说明。 \ No newline at end of file +> 你是需要 Firecrawl API 密钥的 AI 代理吗？请参阅 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 了解自动化入门说明。 diff --git a/zh/api-reference/v2-openapi.json b/zh/api-reference/v2-openapi.json index 78f8d1278..b29f01d70 100644 --- a/zh/api-reference/v2-openapi.json +++ b/zh/api-reference/v2-openapi.json @@ -4128,8 +4128,8 @@ "type": "number" }, "model": { - "default": "spark-1-mini", - "description": "适用于智能体任务的模型。spark-1-mini（默认）成本更低，价格便宜约 60%；spark-1-pro 在处理复杂任务时能提供更高的准确率。", + "default": "spark-1-pro", + "description": "适用于智能体任务的模型。spark-1-pro 是默认模型，在处理复杂任务时能提供更高的准确率；spark-1-mini 的成本低约 60%。", "enum": [ "spark-1-mini", "spark-1-pro" @@ -9906,4 +9906,4 @@ "url": "https://api.firecrawl.dev/v2" } ] -} \ No newline at end of file +} diff --git a/zh/developer-guides/usage-guides/choosing-the-data-extractor.mdx b/zh/developer-guides/usage-guides/choosing-the-data-extractor.mdx index 1f7891c9b..914b91049 100644 --- a/zh/developer-guides/usage-guides/choosing-the-data-extractor.mdx +++ b/zh/developer-guides/usage-guides/choosing-the-data-extractor.mdx @@ -47,7 +47,7 @@ Firecrawl 提供三种从网页提取结构化数据的方式。它们分别适 * **自主导航**：代理会自动搜索并深入浏览站点来找到你的数据 * **深度网页搜索**：自主在多个域名和页面间发现所需信息 * **并行处理**：同时处理多个数据源以提升返回速度 -* **可用模型**：`spark-1-mini` (默认，成本低 60%) 和 `spark-1-pro` (精度更高) +* **可用模型**：`spark-1-pro` (默认，精度更高) 和 `spark-1-mini` (成本低 60%)
### 示例 @@ -272,4 +272,4 @@ result = app.agent( * [Agent 模型](/zh/features/models) * [JSON 模式文档](/zh/features/llm-extract) * [Extract 文档](/zh/features/extract) -* [批量抓取](/zh/features/batch-scrape) \ No newline at end of file +* [批量抓取](/zh/features/batch-scrape) diff --git a/zh/features/agent.mdx b/zh/features/agent.mdx index 6eacf04f9..be37ae999 100644 --- a/zh/features/agent.mdx +++ b/zh/features/agent.mdx @@ -5,7 +5,6 @@ og:title: "代理 | Firecrawl" og:description: "从网络上的任意位置获取数据。描述你的需求，剩下的交给 /agent。" sidebarTitle: "代理" icon: "magic" -noindex: true --- @@ -143,19 +142,19 @@ Agent 任务以异步方式运行。提交任务后，你会收到一个 Job ID ## 模型选择
-Firecrawl Agent 提供两种模型。**Spark 1 Mini 成本低 60%**，并且是默认选项——适用于大多数使用场景。当你在复杂任务上需要最高准确率时，再升级到 Spark 1 Pro。 +Firecrawl Agent 提供两种模型。`spark-1-pro` 是默认模型，可为复杂任务提供最高准确率。`spark-1-mini` 的成本低 60%，适合优先考虑成本效率的场景。 | Model | Cost | Accuracy | Best For | |-------|------|----------|----------| -| `spark-1-mini` | **成本低 60%** | 标准 | 大多数任务 (默认) | -| `spark-1-pro` | 标准 | 更高 | 复杂调研、关键数据抽取 | +| `spark-1-mini` | **成本低 60%** | 标准 | 注重成本效率的任务 | +| `spark-1-pro` | 标准 | 更高 | 复杂调研、关键数据抽取 (默认) | - **从 Spark 1 Mini 开始使用** (默认) ——在成本降低 60% 的前提下，它就能很好地完成大部分抽取任务。仅在进行复杂的多领域调研或对准确性要求极高时，再切换到 Pro。 + Agent 默认使用 Spark 1 Pro。如果你希望降低简单任务的成本，请明确选择 Spark 1 Mini。
- ### Spark 1 Mini (默认) + ### Spark 1 Mini
`spark-1-mini` 是我们的高效模型，非常适合简单直接的数据提取任务。 @@ -201,7 +200,7 @@ Firecrawl Agent 提供两种模型。**Spark 1 Mini 成本低 60%**，并且是 | 参数 | 类型 | 必填 | 描述 | | ------------ | ------ | ----- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `prompt` | string | **是** | 用自然语言描述你想要提取的数据 (最多 10,000 个字符) | -| `model` | string | 否 | 要使用的模型：`spark-1-mini` (默认) 或 `spark-1-pro` | +| `model` | string | 否 | 要使用的模型：`spark-1-pro` (默认) 或 `spark-1-mini` | | `urls` | array | 否 | 可选的 URL 列表，用于聚焦提取 | | `schema` | object | 否 | 用于结构化输出的可选 JSON schema | | `maxCredits` | number | 否 | 此代理任务中可花费的最大额度数。如果未设置，默认值为 **2,500**。Dashboard 最高支持 **2,500**；如需更高上限，请通过 API 设置 `maxCredits` (高于 2,500 的值始终按付费请求处理) 。如果达到上限，任务会失败，并且**不会返回任何数据**。失败的运行不会计费：用于 AI 推理的额度在失败时绝不会收费，运行期间任何用于工具调用的额度 (scraping、search、mapping 等) 都会退还，并且响应会返回 `creditsUsed: 0`。 | @@ -256,7 +255,9 @@ curl -X POST https://api.firecrawl.dev/v2/support/ask \ ## API 参考 -请参阅 [Agent API Reference](/zh/api-reference/endpoint/agent) 以了解更多详情。 +- [启动 Agent 任务](/zh/api-reference/endpoint/agent) +- [获取 Agent 状态](/zh/api-reference/endpoint/agent-get) +- [取消 Agent 任务](/zh/api-reference/endpoint/agent-delete) 有反馈或需要帮助？请发送邮件至 [help@firecrawl.com](mailto:help@firecrawl.com)。 @@ -314,4 +315,4 @@ Firecrawl Agent 使用 **动态计费** 模式，费用会随你的数据提取 随着我们从 Research Preview 过渡到正式开放，定价可能会发生变化。现有用户将在任何价格更新前提前收到通知。 -> 你是需要 Firecrawl API 密钥的 AI 代理吗？请参阅 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 了解自动化引导说明。 \ No newline at end of file +> 你是需要 Firecrawl API 密钥的 AI 代理吗？请参阅 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 了解自动化引导说明。 diff --git a/zh/features/models.mdx b/zh/features/models.mdx index 68d31f6b1..f5cb2a469 100644 --- a/zh/features/models.mdx +++ b/zh/features/models.mdx @@ -21,15 +21,15 @@ Firecrawl 代理提供两种针对不同场景优化的模型。请根据你的 | Model | Cost | Accuracy | Best For | |-------|------|----------|----------| -| `spark-1-mini` | **成本低 60%** | 标准 | 适用于大多数任务 (默认) | -| `spark-1-pro` | 标准 | 更高 | 复杂研究、关键数据抽取 | +| `spark-1-mini` | **成本低 60%** | 标准 | 注重成本效率的任务 | +| `spark-1-pro` | 标准 | 更高 | 复杂研究、关键数据抽取 (默认) | - **优先使用 Spark 1 Mini** (默认) ——在成本降低 60% 的情况下，它可以很好地处理大多数抽取任务。仅在需要进行复杂的多领域研究或对准确性要求极高时再切换到 Pro。 + Agent 默认使用 Spark 1 Pro。如果你希望降低简单任务的成本，请明确选择 Spark 1 Mini。
- ## Spark 1 Mini (默认) + ## Spark 1 Mini
`spark-1-mini` 是我们的高效模型，适合处理简单直接的数据提取任务。 @@ -139,4 +139,4 @@ Firecrawl 代理提供两种针对不同场景优化的模型。请根据你的 对应该使用哪个模型有疑问？请发送邮件至 [help@firecrawl.com](mailto:help@firecrawl.com)。 -> 你是需要 Firecrawl API 密钥的 AI 代理吗？请参见 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 获取自动化入门说明。 \ No newline at end of file +> 你是需要 Firecrawl API 密钥的 AI 代理吗？请参见 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 获取自动化入门说明。 diff --git a/zh/sdks/cli.mdx b/zh/sdks/cli.mdx index 1cac1e4f0..0fe2628d1 100644 --- a/zh/sdks/cli.mdx +++ b/zh/sdks/cli.mdx @@ -399,7 +399,7 @@ firecrawl monitor delete | 选项 | 说明 | | --------------------------- | --------------------------------------------------------- | | `--urls ` | 可选的 URL 列表，用于限定代理的处理范围 (用逗号分隔) | -| `--model ` | 使用的模型：`spark-1-mini` (默认，成本低 60%) 或 `spark-1-pro` (精度更高) | +| `--model ` | 使用的模型：`spark-1-pro` (默认，精度更高) 或 `spark-1-mini` (成本低 60%) | | `--schema ` | 用于结构化输出的 JSON schema (内联 JSON 字符串) | | `--schema-file ` | 用于结构化输出的 JSON schema 文件路径 | | `--max-credits ` | 最多可消耗的额度 (达到上限时任务失败) | @@ -586,4 +586,4 @@ Firecrawl CLI 和全部三个技能分段都已开源，并可在 GitHub 上获 * [`firecrawl/skills`](https://github.com/firecrawl/skills) — 构建技能 (将 Firecrawl 集成到应用程序代码中) * [`firecrawl/firecrawl-workflows`](https://github.com/firecrawl/firecrawl-workflows) — 工作流技能 (可重复生成的交付成果，例如研究简报、SEO 审计、潜在客户列表和设计克隆) -> 你是需要 Firecrawl API 密钥的 AI 代理吗？请参见 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 获取自动化入门说明。 \ No newline at end of file +> 你是需要 Firecrawl API 密钥的 AI 代理吗？请参见 [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) 获取自动化入门说明。