Deterministic evaluation harness and task suite for GAIA (General AI Assistants) multi-modal benchmark.
-
Updated
Oct 4, 2026 - Python
#
gaia-benchmark
Here are 12 public repositories matching this topic...
PECS: 基于 LangGraph 的四角色多智能体任务求解框架（Planner/Executor/Critic/Synthesizer）。WebShop 真实环境 +25pp (25% vs 0%)；GAIA 官方 53 题 26.4% vs ReAct 24.5%（McNemar 不显著）。含 AST 沙箱、50000 token 硬预算、FastAPI 限流/混沌/CI/Prometheus。
python flask multi-agent ai-agents cost-optimization ablation-study ai-agent llm-agent langgraph deepseek-api token-optimization gaia-benchmark plan-execute-reflect agentbench ast-sandbox
-
Updated
Oct 14, 2026 - Python
Repository with a LangGraph general purpose AI Agent for final assignment of HuggingFace Agents course.
-
Updated
May 20, 2025 - Python
Tetrahedral geometry attention for GAIA benchmarks - 64-point synergetic transformer
-
Updated
Jan 14, 2026 - Python
Streamlit tool for benchmarking OpenAI models against GAIA test cases with feedback-driven re-evaluation
-
Updated
Oct 11, 2024
Final assignment for Hugging Face's Agents Course (Unit 4) — a GAIA benchmark agent built with smolagents CodeAgent + GPT-4o, deployed on Hugging Face Spaces to answer and auto-submit Level-1 GAIA validation questions via their scoring API.
ai gradio ai-agents huggingface-spaces gpt-4o agentic-ai smolagents gaia-benchmark huggingface-agents-course
-
Updated
Aug 5, 2026 - Python
🕵🏻♂️ LangGraph-powered AI agent tackling GAIA Benchmark tasks, built for Hugging Face Agents Course.
python gradio ruff ai-agents uv huggingface groq llm huggingface-spaces react-agent langgraph tavily gaia-benchmark hf-agent
-
Updated
Jun 17, 2025 - Python
smolagents CodeAgent for the GAIA final exam of the Hugging Face AI Agents Course, on Groq's free tier
-
Updated
Oct 7, 2026 - Python
Tool-using AI agent built with LangGraph + Gemini. 85% (17/20) on GAIA level-1 validation questions. Final project of the Hugging Face AI Agents course.
-
Updated
Oct 6, 2026 - Python
A general-purpose AI agent powered by Claude and the Agent SDK — web search, file processing, vision, audio transcription, precise math, and multi-agent orchestration from your terminal.
-
Updated
Apr 2, 2026 - Python
Supervisor multi-agent system for autonomous research, reasoning and code execution. Built on LangGraph. Targeting GAIA.
python ai multi-agent gaia agents observability autonomous-agents ai-agents tool-use llm langchain langsmith react-agent langgraph gaia-benchmark
-
Updated
Sep 28, 2026 - Python
An AI agent built from scratch in Java, chapter by chapter along Manning's "Build an AI Agent (From Scratch)". ReAct loop, tools, MCP, vector search and GAIA evaluation are hand-written; Spring AI is only the layer to OpenAI, Anthropic and Gemini.
-
Updated
Sep 29, 2026 - Java
Add this topic to your repo
To associate your repository with the gaia-benchmark topic, visit your repo's landing page and select "manage topics."