-
Notifications
You must be signed in to change notification settings - Fork 39
Pull requests: fw-ai/cookbook
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(infra): skip pod-identity wait on same-trainer reattach
#319
opened Apr 10, 2026 by
Hecate0821
Contributor
Loading…
3 of 4 tasks
fix(infra): detect UPDATING→READY in reattach settle to avoid timeout
#318
opened Apr 10, 2026 by
Hecate0821
Contributor
Loading…
2 tasks
feat(checkpoint): validate checkpoint entries before resume
#314
opened Apr 9, 2026 by
hershalb
Contributor
Loading…
14 of 23 tasks
feat(training): add accelerator_type filtering to training shape selection
#308
opened Apr 7, 2026 by
hershalb
Contributor
Loading…
Add KV-cache key compression reproducible comparison
#303
opened Apr 6, 2026 by
yi-fireworks
Loading…
4 of 5 tasks
feat: LoRA self-reference RL — use policy trainer as KL reference
#299
opened Apr 6, 2026 by
mayinghan
Contributor
Loading…
[codex] Fix training SDK smoke compatibility
#292
opened Apr 2, 2026 by
benjibc
Contributor
Loading…
docs(training): update cookbook install guidance
#289
opened Apr 1, 2026 by
benjibc
Contributor
Loading…
3 tasks done
feat: IGPO training recipe and multi-hop QA example
#284
opened Mar 31, 2026 by
morgendave
Loading…
6 tasks done
Reduce cleanup grace period from 30s to 5s
#267
opened Mar 26, 2026 by
mayinghan
Contributor
Loading…
2 tasks
feat(training): route rl loop references through training sessions
#245
opened Mar 20, 2026 by
renfeichen-fw
Contributor
•
Draft
feat(training): async RL loop with AReaL-style streaming pipeline
#243
opened Mar 19, 2026 by
Hecate0821
Contributor
Loading…
4 tasks done
fix(training): harden cleanup and improve SFT observability
#228
opened Mar 17, 2026 by
mayinghan
Contributor
Loading…
fix: point deepmath launcher at validated grpo path
#214
opened Mar 15, 2026 by
benjibc
Contributor
Loading…
fix(training): separate gspo and gspo-token losses
#212
opened Mar 15, 2026 by
benjibc
Contributor
Loading…
2 of 3 tasks
fix(rl): reuse policy trainer for lora grpo references
#211
opened Mar 15, 2026 by
benjibc
Contributor
Loading…
2 of 3 tasks
fix(frozen-lake): harden gateway-backed smoke path
#206
opened Mar 13, 2026 by
benjibc
Contributor
Loading…
1 task done
Update run.sh for qwen3-32b model configuration
#203
opened Mar 13, 2026 by
mayinghan
Contributor
Loading…
fix: pass fw_api_key in ORPO/DPO recipes and guard wandb entity
#197
opened Mar 12, 2026 by
renfeichen-fw
Contributor
Loading…
2 of 4 tasks
Previous Next
ProTip!
Adding no:label will show everything without a label.