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 }}
sgl-project
/
SpecForge
Public
Notifications
You must be signed in to change notification settings
Fork
358
Star
1.2k
Code
Issues
75
Pull requests
125
Discussions
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Projects
Security and quality
Insights
Actions: sgl-project/SpecForge
Actions
All workflows
Workflows
CodeQL
CodeQL
Copilot
Copilot
Copilot code review
Copilot code review
Dependency Graph
Dependency Graph
Lint
Lint
pages-build-deployment
pages-build-deployment
PR Test
PR Test
Publish to PyPI
Publish to PyPI
Release Documentation
Release Documentation
Show more workflows...
Management
Caches
Deployments
All workflows
All workflows
Actions
Loading...
Loading
Sorry, something went wrong.
Uh oh!
There was an error while loading.
Please reload this page
.
will be ignored since log searching is not yet available
Showing runs from all workflows
will be ignored since log searching is not yet available
2,500+ workflow runs
2,500+ workflow runs
Workflow
Filter by Workflow
Sorry, something went wrong.
Filter
Loading
Sorry, something went wrong.
No matching workflows.
Event
Filter by Event
Sorry, something went wrong.
Filter
Loading
Sorry, something went wrong.
No matching events.
Status
Filter by Status
Sorry, something went wrong.
Filter
Loading
Sorry, something went wrong.
No matching statuses.
Branch
Filter by Branch
Sorry, something went wrong.
Filter
Loading
Sorry, something went wrong.
No matching branches.
Actor
Filter by Actor
Sorry, something went wrong.
Filter
Loading
Sorry, something went wrong.
No matching users.
perf(dflash): compiled create_block_mask / flex_attention wrappers [3/3]
Lint
#2476:
Pull request
#936
opened by
maocheng23
54s
dflash-compiled-block-mask
dflash-compiled-block-mask
54s
View #936
View workflow file
feat(dflash): share DFlash masks with inference, default masks for spec_generate [2/3]
Lint
#2475:
Pull request
#935
opened by
maocheng23
49s
dflash-inference-masks
dflash-inference-masks
49s
View #935
View workflow file
fix(dflash): honor config.is_causal in DFlash training masks [1/3]
Lint
#2474:
Pull request
#934
opened by
maocheng23
42s
dflash-is-causal-min
dflash-is-causal-min
42s
View #934
View workflow file
fix(dflash): honor config.is_causal in DFlash training masks [1/3]
PR Test
#1928:
Pull request
#934
opened by
maocheng23
In progress
dflash-is-causal-min
dflash-is-causal-min
In progress
View #934
View workflow file
PR #934
CodeQL
#611:
by
maocheng23
46s
refs/pull/934/head
refs/pull/934/head
46s
fix(data): supervise historical Qwen thinking turns
PR Test
#1927:
Pull request
#933
opened by
cheater6666
Action required
cheater6666:fix/qwen-history-loss-mask
cheater6666:fix/qwen-history-loss-mask
Action required
View #933
View workflow file
fix(data): supervise historical Qwen thinking turns
Lint
#2473:
Pull request
#933
opened by
cheater6666
Action required
cheater6666:fix/qwen-history-loss-mask
cheater6666:fix/qwen-history-loss-mask
Action required
View #933
View workflow file
(fsdp2 dev.) feat(moe): expert parallelism for MoE drafters on the FSDP2 backend (stacked on #915 + #931)
Lint
#2472:
Pull request
#932
opened by
yushengsu-thu
Action required
yushengsu-thu:moe-ep-fsdp2
yushengsu-thu:moe-ep-fsdp2
Action required
View #932
View workflow file
(fsdp2 dev.) feat(moe): expert parallelism for MoE drafters on the FSDP2 backend (stacked on #915 + #931)
PR Test
#1926:
Pull request
#932
opened by
yushengsu-thu
Action required
yushengsu-thu:moe-ep-fsdp2
yushengsu-thu:moe-ep-fsdp2
Action required
View #932
View workflow file
exp - feat(moe): qwen3 preset and the Qwen3.8-27B DFlash2 MoE ablation arm (stacked on #811/#812)
Lint
#2471:
Pull request
#930
reopened by
yushengsu-thu
-1s
yushengsu-thu:feat/qwen3.8-27b-dflash2-moe
yushengsu-thu:feat/qwen3.8-27b-dflash2-moe
-1s
View #930
View workflow file
exp - feat(moe): qwen3 preset and the Qwen3.8-27B DFlash2 MoE ablation arm (stacked on #811/#812)
PR Test
#1925:
Pull request
#930
reopened by
yushengsu-thu
Action required
yushengsu-thu:feat/qwen3.8-27b-dflash2-moe
yushengsu-thu:feat/qwen3.8-27b-dflash2-moe
Action required
View #930
View workflow file
exp.[rebase] MoE FFN for DFlash-family drafters (#811/#812) on current main, to run the DeepSeek-V4-Flash MoE arm
Lint
#2470:
Pull request
#931
opened by
yushengsu-thu
Action required
yushengsu-thu:moe-stack-rebased
yushengsu-thu:moe-stack-rebased
Action required
View #931
View workflow file
exp.[rebase] MoE FFN for DFlash-family drafters (#811/#812) on current main, to run the DeepSeek-V4-Flash MoE arm
PR Test
#1924:
Pull request
#931
opened by
yushengsu-thu
Action required
yushengsu-thu:moe-stack-rebased
yushengsu-thu:moe-stack-rebased
Action required
View #931
View workflow file
exp - feat(moe): qwen3 preset and the Qwen3.8-27B DFlash2 MoE ablation arm (stacked on #811/#812)
PR Test
#1923:
Pull request
#930
opened by
yushengsu-thu
18m 38s
yushengsu-thu:feat/qwen3.8-27b-dflash2-moe
yushengsu-thu:feat/qwen3.8-27b-dflash2-moe
18m 38s
View #930
View workflow file
exp - feat(moe): qwen3 preset and the Qwen3.8-27B DFlash2 MoE ablation arm (stacked on #811/#812)
Lint
#2469:
Pull request
#930
opened by
yushengsu-thu
18m 38s
yushengsu-thu:feat/qwen3.8-27b-dflash2-moe
yushengsu-thu:feat/qwen3.8-27b-dflash2-moe
18m 38s
View #930
View workflow file
(fsdp2 dev.) - follow-up : Add training.static_shape_buckets: a few static lengths for compiled blocks
Lint
#2468:
Pull request
#929
synchronize by
yushengsu-thu
Action required
yushengsu-thu:fsdp2/shape-buckets
yushengsu-thu:fsdp2/shape-buckets
Action required
View #929
View workflow file
(fsdp2 dev.) - follow-up : Add training.static_shape_buckets: a few static lengths for compiled blocks
PR Test
#1922:
Pull request
#929
synchronize by
yushengsu-thu
Action required
yushengsu-thu:fsdp2/shape-buckets
yushengsu-thu:fsdp2/shape-buckets
Action required
View #929
View workflow file
(fsdp2 dev. perf) Add training.fp8_linear: torchao Float8Linear with float8 all-gather
PR Test
#1921:
Pull request
#923
synchronize by
yushengsu-thu
Action required
yushengsu-thu:fsdp2/fp8-linear
yushengsu-thu:fsdp2/fp8-linear
Action required
View #923
View workflow file
(fsdp2 dev. perf) Add training.fp8_linear: torchao Float8Linear with float8 all-gather
Lint
#2467:
Pull request
#923
synchronize by
yushengsu-thu
Action required
yushengsu-thu:fsdp2/fp8-linear
yushengsu-thu:fsdp2/fp8-linear
Action required
View #923
View workflow file
(fsdp2 dev. perf) Add training.compile_blocks: per-block torch.compile before fully_shard
PR Test
#1920:
Pull request
#922
synchronize by
yushengsu-thu
Action required
yushengsu-thu:fsdp2/compile-blocks
yushengsu-thu:fsdp2/compile-blocks
Action required
View #922
View workflow file
(fsdp2 dev. perf) Add training.compile_blocks: per-block torch.compile before fully_shard
Lint
#2466:
Pull request
#922
synchronize by
yushengsu-thu
Action required
yushengsu-thu:fsdp2/compile-blocks
yushengsu-thu:fsdp2/compile-blocks
Action required
View #922
View workflow file
(fsdp2 dev.) - follow-up : Add training.static_shape_buckets: a few static lengths for compiled blocks
PR Test
#1919:
Pull request
#929
opened by
yushengsu-thu
Action required
yushengsu-thu:fsdp2/shape-buckets
yushengsu-thu:fsdp2/shape-buckets
Action required
View #929
View workflow file
(fsdp2 dev.) - follow-up : Add training.static_shape_buckets: a few static lengths for compiled blocks
Lint
#2465:
Pull request
#929
opened by
yushengsu-thu
Action required
yushengsu-thu:fsdp2/shape-buckets
yushengsu-thu:fsdp2/shape-buckets
Action required
View #929
View workflow file
(fsdp2 dev. perf) Add sequence packing for EAGLE3, DFlash, and DFlash2
PR Test
#1918:
Pull request
#928
synchronize by
yushengsu-thu
-1s
yushengsu-thu:codex/sequence-packing
yushengsu-thu:codex/sequence-packing
-1s
View #928
View workflow file
(fsdp2 dev. perf) Add sequence packing for EAGLE3, DFlash, and DFlash2
Lint
#2464:
Pull request
#928
synchronize by
yushengsu-thu
-1s
yushengsu-thu:codex/sequence-packing
yushengsu-thu:codex/sequence-packing
-1s
View #928
View workflow file
Previous
1
2
3
4
5
…
99
100
101
Next
You can’t perform that action at this time.