Skip to content
Navigation Menu
Sign in
Appearance settings
Platform
AI CODE CREATION
GitHub Copilot
Write better code with AI
GitHub Copilot app
Direct agents from issue to merge
MCP Registry
Integrate external tools
DEVELOPER WORKFLOWS
Actions
Automate any workflow
Codespaces
Instant dev environments
Issues
Plan and track work
Code Review
Manage code changes
Code Quality
Enforce quality at merge
APPLICATION SECURITY
GitHub Advanced Security
Find and fix vulnerabilities
Code security
Secure your code as you build
Secret protection
Stop leaks before they start
EXPLORE
Why GitHub
Documentation
Blog
Changelog
Marketplace
View all features
Solutions
BY COMPANY SIZE
Enterprises
Small and medium teams
Startups
Nonprofits
BY USE CASE
App Modernization
DevSecOps
DevOps
CI/CD
View all use cases
BY INDUSTRY
Healthcare
Financial services
Manufacturing
Government
View all industries
View all solutions
Resources
EXPLORE BY TOPIC
AI
Software Development
DevOps
Security
View all topics
EXPLORE BY TYPE
Customer stories
Events & webinars
Ebooks & reports
Business insights
GitHub Skills
SUPPORT & SERVICES
Documentation
Customer support
Community forum
Trust center
Partners
View all resources
Open Source
COMMUNITY
GitHub Sponsors
Fund open source developers
PROGRAMS
Security Lab
Maintainer Community
GitHub Stars
Archive Program
REPOSITORIES
Topics
Trending
Collections
Enterprise
ENTERPRISE SOLUTIONS
Enterprise platform
AI-powered developer platform
AVAILABLE ADD-ONS
GitHub Advanced Security
Enterprise-grade security features
Copilot for Business
Enterprise-grade AI features
Premium Support
Enterprise-grade 24/7 support
Pricing
Search
/
Sign in
Sign up
Appearance settings
You signed in with another tab or window.
Reload
to refresh your session.
You signed out in another tab or window.
Reload
to refresh your session.
You switched accounts on another tab or window.
Reload
to refresh your session.
Dismiss alert
{{ message }}
bsreecharanreddy
dispatch
Repository navigation
Code
Issues
Pull requests
1
(1)
Actions
Projects
Security and quality
Insights
More
items
Phase 5a: int8 weight-only quantization
- #5
#5
Merged
bsreecharanreddy
merged 18 commits into
main
bsreecharanreddy/dispatch:main
from
phase-5a-quantization
bsreecharanreddy/dispatch:phase-5a-quantization
Copy head branch name to clipboard
Sep 17, 2026
Conversation
Commits
18
(18)
Checks
Files changed
Merged
Phase 5a: int8 weight-only quantization
#5
bsreecharanreddy
merged 18 commits into
main
bsreecharanreddy/dispatch:main
from
phase-5a-quantization
bsreecharanreddy/dispatch:phase-5a-quantization
Copy head branch name to clipboard
Commits
Commits on Sep 17, 2026
docs: add phase 5a quantization design doc
Show description for dd5bb5d
bsreecharanreddy
committed
dd5bb5d
View commit details
Copy full SHA for dd5bb5d
Browse repository at this point
docs: add phase 5a quantization implementation plan
bsreecharanreddy
committed
0371da4
View commit details
Copy full SHA for 0371da4
Browse repository at this point
style: ruff-format the plan doc's embedded code blocks
bsreecharanreddy
committed
49da8b7
View commit details
Copy full SHA for 49da8b7
Browse repository at this point
docs: fix plan's import-placement instructions to avoid E402
bsreecharanreddy
committed
1acc7c0
View commit details
Copy full SHA for 1acc7c0
Browse repository at this point
feat: add int8 per-channel weight quantization primitives
bsreecharanreddy
committed
4ce0710
View commit details
Copy full SHA for 4ce0710
Browse repository at this point
fix: use input dtype for NaN-guard clamp floor in quantize_per_channel_int8
bsreecharanreddy
committed
1913993
View commit details
Copy full SHA for 1913993
Browse repository at this point
feat: add quantized MoE forward path and memory-footprint measurement
bsreecharanreddy
committed
c77e4a9
View commit details
Copy full SHA for c77e4a9
Browse repository at this point
feat: add Triton int8 grouped-GEMM kernel and its backend resolver
bsreecharanreddy
committed
b231aa8
View commit details
Copy full SHA for b231aa8
Browse repository at this point
fix: keep int8 kernel's tl.dot on native tensor-core precision, validate scale shape
Show description for c87c332
bsreecharanreddy
committed
c87c332
View commit details
Copy full SHA for c87c332
Browse repository at this point
feat: add patch_moe_infer_quantized, sharing layer-iteration with patch_moe_infer
bsreecharanreddy
committed
20d61ff
View commit details
Copy full SHA for 20d61ff
Browse repository at this point
feat: wire --moe-kernel quantized through patch_moe_infer_quantized
bsreecharanreddy
committed
0fe78bf
View commit details
Copy full SHA for 0fe78bf
Browse repository at this point
fix: free original bf16 expert weights in patch_moe_infer_quantized
Show description for c75da79
bsreecharanreddy
committed
c75da79
View commit details
Copy full SHA for c75da79
Browse repository at this point
docs: record Phase 5a int8 quantized kernel outcome
Show description for 60c2ebe
bsreecharanreddy
committed
60c2ebe
View commit details
Copy full SHA for 60c2ebe
Browse repository at this point
docs: refresh README with almanac-style sections and Phase 5a results
Show description for 7e98bcd
bsreecharanreddy
committed
7e98bcd
View commit details
Copy full SHA for 7e98bcd
Browse repository at this point
docs: fix two deferred cosmetic findings from the task-review loop
Show description for ac46971
bsreecharanreddy
committed
ac46971
View commit details
Copy full SHA for ac46971
Browse repository at this point
fix: compute int8 quantization arithmetic in float32, fix fp16 clamp cliff
Show description for 589834c
bsreecharanreddy
committed
589834c
View commit details
Copy full SHA for 589834c
Browse repository at this point
docs: address final-review documentation and cleanliness findings
Show description for 1ef2ded
bsreecharanreddy
committed
1ef2ded
View commit details
Copy full SHA for 1ef2ded
Browse repository at this point
docs: fix CHANGELOG chronological order and a stale test count
Show description for 2f70747
bsreecharanreddy
committed
2f70747
View commit details
Copy full SHA for 2f70747
Browse repository at this point
You can’t perform that action at this time.