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 }}
kmbandy
/
llama.cpp
Public
forked from
ggml-org/llama.cpp
Notifications
You must be signed in to change notification settings
Fork
0
Star
5
Code
Issues
0
Pull requests
0
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Actions
Projects
Security and quality
Insights
Consolidate gfx803 paged/turbo4 + LFM2.5 work onto fork master
- #14
#14
Merged
kmbandy
merged 7 commits into
master
kmbandy/llama.cpp:master
from
fix/lfm25-paged-turbo4-gfx803
kmbandy/llama.cpp:fix/lfm25-paged-turbo4-gfx803
Copy head branch name to clipboard
Jun 26, 2026
Conversation
Commits
7
(7)
Checks
Files changed
Merged
Consolidate gfx803 paged/turbo4 + LFM2.5 work onto fork master
#14
kmbandy
merged 7 commits into
master
kmbandy/llama.cpp:master
from
fix/lfm25-paged-turbo4-gfx803
kmbandy/llama.cpp:fix/lfm25-paged-turbo4-gfx803
Copy head branch name to clipboard
Commits
Commits on Jun 15, 2026
cuda/paged-attn: portable non-WMMA tile kernel + fix avg_q_len grid sizing
Show description for d00fea0
kmbandy
and
claude
committed
d00fea0
View commit details
Copy full SHA for d00fea0
Browse repository at this point
docs: paged-attn debug notes for 2026-06-15 session
Show description for 1d75aee
kmbandy
and
claude
committed
1d75aee
View commit details
Copy full SHA for 1d75aee
Browse repository at this point
Commits on Jun 16, 2026
cuda/paged-attn: fix gfx803 decode under-utilization (CHUNK_KV + size-by-actual)
Show description for 7e5d23d
kmbandy
and
claude
committed
7e5d23d
View commit details
Copy full SHA for 7e5d23d
Browse repository at this point
cuda/mmq: route gfx80x large-batch GEMM to dequant+hipBLAS (MAD-301B)
Show description for 9fbeba2
kmbandy
and
claude
committed
9fbeba2
View commit details
Copy full SHA for 9fbeba2
Browse repository at this point
cuda/paged-attn: portable FMA multi-warp tile kernel for gfx803 (MAD-301C Lever A)
Show description for 60dc49a
kmbandy
and
claude
committed
60dc49a
View commit details
Copy full SHA for 60dc49a
Browse repository at this point
cuda/paged-attn: native head_dim-64 turbo4 KV (GGML_TYPE_TURBO4_64, MAD-301C Lever B)
Show description for 5850420
kmbandy
and
claude
committed
5850420
View commit details
Copy full SHA for 5850420
Browse repository at this point
Commits on Jun 20, 2026
chat: fix LFM2.5 reasoning bleed when opening <think> is omitted
Show description for bf90040
kmbandy
and
claude
committed
bf90040
View commit details
Copy full SHA for bf90040
Browse repository at this point
You can’t perform that action at this time.
