-
Notifications
You must be signed in to change notification settings - Fork 577
Pull requests: rllm-org/rllm
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(fireworks): retry sampling timeouts instead of failing after one attempt
#689
opened Jun 24, 2026 by
jeffreysijuntan
Contributor
Loading…
feat(fireworks): fold ECHO env loss into the client loss path (single pass)
#688
opened Jun 24, 2026 by
jeffreysijuntan
Contributor
•
Draft
feat(renderers): native renderer layer + DeepSeek-V4 rollout & cumulative-mode support
#683
opened Jun 22, 2026 by
jeffreysijuntan
Contributor
Loading…
feat(cli): eval → curate → SFT loop + unified SFT trainer (tinker/fireworks)
#673
opened Jun 20, 2026 by
jeffreysijuntan
Contributor
Loading…
fix(data): serialize extra_info as JSON string in verl postprocessing
#603
opened May 31, 2026 by
bingshao333
Loading…
[wip] refactor render and parse
#594
opened May 22, 2026 by
kylemontgomery1
Collaborator
•
Draft
14 tasks
feat(parser): renderers-based parser backend with selectable parser_backend
#591
opened May 21, 2026 by
listar2000
Collaborator
•
Draft
1 of 3 tasks
refactor(unified_trainer): extract step merging into a shared backend-agnostic module
#576
opened May 10, 2026 by
listar2000
Collaborator
Loading…
2 tasks done
feat(verl): patch zmq IPC id to depend on job id (volcengine/verl#6246)
#569
opened May 7, 2026 by
listar2000
Collaborator
•
Draft
4 of 5 tasks
feat(console): operator UI mounted on the gateway; retire visualizer.py
#558
opened May 5, 2026 by
jeffreysijuntan
Contributor
Loading…
3 tasks done
feat(harnesses): opencode/mini-swe-agent/oracle + Harbor 0.5 + --runtime flag
#557
opened May 5, 2026 by
jeffreysijuntan
Contributor
Loading…
3 tasks done
feat(eval): drop LiteLLM, in-process gateway + tunnel + cleanup
#556
opened May 5, 2026 by
jeffreysijuntan
Contributor
Loading…
3 tasks done
refactor(engine): introduce FlowEngine base, rename WorkflowEngine
#555
opened May 5, 2026 by
listar2000
Collaborator
•
Draft
4 of 14 tasks
feat(model-gateway): upstream-proxy mode + run lifecycle
#553
opened May 5, 2026 by
jeffreysijuntan
Contributor
Loading…
2 tasks done
feat(model-gateway): X-RLLM-* headers + inbound bearer auth
#552
opened May 5, 2026 by
jeffreysijuntan
Contributor
Loading…
3 tasks done
feat(model-gateway): trace store schema v2
#551
opened May 5, 2026 by
jeffreysijuntan
Contributor
Loading…
2 of 3 tasks
fix(verl): build transform attention masks from sequence lengths
#517
opened Apr 29, 2026 by
JasonWei05
Collaborator
Loading…
3 of 14 tasks
fix(verl): preserve multi-turn tool-call prefix extension for math tool agent for Qwen 3 models
#516
opened Apr 29, 2026 by
JasonWei05
Collaborator
•
Draft
6 of 14 tasks
fix: unified async trainer with verl backend
#493
opened Apr 6, 2026 by
yifannnwu
Contributor
Loading…
1 task done
refactor: replace bypass_render_with_parser with TinkerChatTemplateParser
#489
opened Apr 6, 2026 by
listar2000
Collaborator
•
Draft
2 of 3 tasks
Add strict DPO objective plumbing and preference-pair groundwork
#477
opened Apr 2, 2026 by
taivu1998
Contributor
Loading…
Add early-finalize continuation for truncated reasoning rollouts
#475
opened Apr 2, 2026 by
taivu1998
Contributor
Loading…
Added adapator layers for to-be-deprecated AgentWorkflowEngine and AgentExecutionEngine
#413
opened Mar 3, 2026 by
boredbichon67
Contributor
Loading…
Previous Next
ProTip!
no:milestone will show everything without a milestone.