Skip to content
Navigation Menu
Toggle navigation
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
New
Integrate external tools
DEVELOPER WORKFLOWS
Actions
Automate any workflow
Codespaces
Instant dev environments
Issues
Plan and track work
Code Review
Manage code changes
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
Accelerator
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 or jump to...
Search code, repositories, users, issues, pull requests...
Search syntax tips
Provide feedback
Saved searches
Use saved searches to filter your results more quickly
Sign in
Sign up
Appearance settings
Resetting focus
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 }}
Uh oh!
There was an error while loading.
Please reload this page
.
Tencent-Hunyuan
/
UniRL
Public
Notifications
You must be signed in to change notification settings
Fork
48
Star
736
Code
Issues
11
Pull requests
17
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Actions
Projects
Security and quality
Insights
Commits
Branch selector
main
User selector
All users
All time
Commit history
Commits on Jun 30, 2026
perf(sd3): batched-step replay for PE diffusion training (ratio=1 exact) (#144)
Show description for 3351375
CjhHa1
and
Ideny42
authored
3351375
Copy full SHA for 3351375
refactor(types): remove dead trajectory_store, relocate compute_trajectory_positions (#109)
Show description for 87618a5
celve
and
CjhHa1
authored
87618a5
Copy full SHA for 87618a5
feat(refl): add DRaFT-K differentiable-reward backprop trainer (#120)
celve
authored
005fa4a
Copy full SHA for 005fa4a
docs(reward): clarify HPSv3++ score_scale is training-invariant; default to raw mu (#146)
leviking98z-rgb
authored
53bb69f
Copy full SHA for 53bb69f
Commits on Jun 29, 2026
build(deps): raise bagel transformers dependency to 5.6 to match engine stacks (#119)
Show description for 153c013
CjhHa1
and
zzhuoxin1508
authored
153c013
Copy full SHA for 153c013
feat(reward): add HPSv3++ reward scorer and SD3.5 recipe (#142)
Show description for c18873d
haonan3
and
leviking98z-rgb
authored
c18873d
Copy full SHA for c18873d
fix(reward): call HPSv3 reward() by keyword; add SD3.5 recipe (#143)
haonan3
authored
34c6a2e
Copy full SHA for 34c6a2e
feat(trainer): semi-independent reward on a dedicated GPU slab (#134)
Show description for 55c1a9a
haonan3
authored
55c1a9a
Copy full SHA for 55c1a9a
Commits on Jun 27, 2026
fix(reward): reject explicit cuda:<index> in local reward device resolver (#125)
Show description for a8a6ef1
xshrz
and
leviking98z-rgb
authored
a8a6ef1
Copy full SHA for a8a6ef1
fix(ltx2.3): make audio x_T driver-authoritative (salted sibling of video recipe) (#126)
Show description for e728247
leviking98z-rgb
authored
e728247
Copy full SHA for e728247
docs: bagel support fix in model support table(#133)
celve
authored
b31a3cf
Copy full SHA for b31a3cf
feat: add WAN 2.2 V2V trainside support (#113)
Show description for 3e709e7
xshrz
and
leviking98z-rgb
authored
3e709e7
Copy full SHA for 3e709e7
docs: note LTX-Video-2.3 T2AV support in model table (#132)
Show description for 5107427
KemingWu
authored
5107427
Copy full SHA for 5107427
Commits on Jun 26, 2026
fix(bagel): replace removed get_diffusion_params with modality-keyed dict access (#130)
Show description for cc1caed
NancyFyong
authored
cc1caed
Copy full SHA for cc1caed
perf(vllm-omni): group SD3 and Qwen-Image prompt rollouts (#117)
Show description for 1e3db3e
CjhHa1
and
Jayce-Ping
committed
1e3db3e
Copy full SHA for 1e3db3e
feat: integrate BAGEL-7B-MoT into vLLM-Omni rollout engine (#89)
zzhuoxin1508
authored
ae2693a
Copy full SHA for ae2693a
feat(ltx2.3): add LTX-2.3 T2AV support with joint audio+video SDE policy (#115)
Show description for caa6656
KemingWu
and
leviking98z-rgb
authored
caa6656
Copy full SHA for caa6656
Commits on Jun 25, 2026
chore(gitignore): keep ignoring reward/local bytecode caches (#123)
Show description for 57d06bd
CjhHa1
and
celve
authored
57d06bd
Copy full SHA for 57d06bd
Commits on Jun 24, 2026
build(deps): unblock the sglang install by overriding diffusers>=0.38 (#118)
CjhHa1
authored
35fb66c
Copy full SHA for 35fb66c
fix(datasets): write pickscore test split to relative path (#106)
Show description for 70056e3
3 people
authored
70056e3
Copy full SHA for 70056e3
Commits on Jun 23, 2026
docs(flowdppo): list official HF checkpoints, clarify curve is SD3.5-medium (#114)
haonan3
authored
f5f8ae0
Copy full SHA for f5f8ae0
feat: qwen image rollout engines( sglang & vllm) (#34)
Show description for 3cb60f6
CjhHa1
authored
3cb60f6
Copy full SHA for 3cb60f6
refactor(train): unify FSDP checkpoint meta-checks into one guard (#111)
Show description for 114b7ab
CjhHa1
and
zzhuoxin1508
authored
114b7ab
Copy full SHA for 114b7ab
fix(qwen3): disable cuDNN SDPA in AR replay to avoid NaN grads (#110)
Show description for fb5ad62
CjhHa1
authored
fb5ad62
Copy full SHA for fb5ad62
Commits on Jun 22, 2026
Update citation format in README.md (#104)
Show description for f4ce93e
zhouxiangxin1998
and
CjhHa1
authored
f4ce93e
Copy full SHA for f4ce93e
docs: note Bagel supports Text + Image → Image in model table (#107)
Show description for 643367c
KemingWu
and
CjhHa1
authored
643367c
Copy full SHA for 643367c
Commits on Jun 19, 2026
feat(ltx2): add LTX-Video-2 model bundle + T2V trainside recipes (#91)
KemingWu
authored
d74d4fb
Copy full SHA for d74d4fb
feat(bagel): it2i editreward RL — recipe + VAE decode fix + diffusion eval (#88)
heguangxin
authored
8067dc4
Copy full SHA for 8067dc4
feat(bagel): trainside it2t (understanding) GRPO RL on BagelARStage (#100)
heguangxin
authored
ebf0546
Copy full SHA for ebf0546
feat(pe): prompt-level (root) GRPO grouping for the diffusion track (#80)
Jayce-Ping
authored
e151bb6
Copy full SHA for e151bb6
feat(pe): inject pe_instruction + strip pe_marker in trainside PEPipeline (#96)
Jayce-Ping
authored
e157b50
Copy full SHA for e157b50
feat(qwen-vl): packed-varlen replay + tf-5.x get_rope_index compat (#40) (#76)
haonan3
authored
8df46f7
Copy full SHA for 8df46f7
feat(veomni): Ulysses sequence parallelism for the VeOmni backend (AR + diffusion) (#71)
celve
authored
c3b9563
Copy full SHA for c3b9563
feat(rollout): canonical sglang LLM/VLM rollout engine, retire sglang_llm (#74)
celve
authored
1b8ee14
Copy full SHA for 1b8ee14
refactor(sampling): modality-keyed sampling dict in RolloutReq, drop ComposedSamplingParams (#70)
celve
authored
30654a8
Copy full SHA for 30654a8
Previous
Next
You can’t perform that action at this time.
