Skip to content

Pull requests: PrimeIntellect-ai/prime-rl

New pull request New
118 Open 2,435 Closed
118 Open 2,435 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[codex] productionize sampled-logprob fast path
#2891 opened Jun 27, 2026 by joanvelja • Draft
refactor: remove "auto" option from fused_lm_head_token_chunk_size
#2890 opened Jun 26, 2026 by samsja Member Loading…
feat(trainer): pack RL multimodal samples
#2889 opened Jun 26, 2026 by hubert-marek Contributor • Draft
8
exp: r2e-gym ablation (vm sandboxes)
#2888 opened Jun 26, 2026 by S1ro1 Collaborator • Draft
feat(dockerfile): opt-in disagg extras build (INCLUDE_DISAGG_EXTRAS)
#2886 opened Jun 26, 2026 by JannikSt Member • Draft
feat(dockerfile): opt-in NIXL-from-source build (INCLUDE_NIXL_FROM_SOURCE)
#2885 opened Jun 26, 2026 by JannikSt Member Loading…
6
feat: adapt orchestrator to pull-based env server (lazy/infinite tasksets)
#2882 opened Jun 25, 2026 by mikasenghaas Member • Draft
Add penalties for turns and input tokens
#2869 opened Jun 24, 2026 by faresobeid Contributor Loading…
2
feat: add orchestrator debug mode (no-inference, no-trainer)
#2867 opened Jun 24, 2026 by samsja Member Loading…
7
feat: configurable per-component env vars for RL runs
#2863 opened Jun 24, 2026 by mikasenghaas Member • Draft
exp: scaleswe ablation
#2862 opened Jun 24, 2026 by S1ro1 Collaborator • Draft
feat(orchestrator): route trace algorithms through loss channels
#2859 opened Jun 23, 2026 by willccbb Member • Draft
4
feat(orchestrator): add trace support on algorithm abstraction
#2857 opened Jun 23, 2026 by willccbb Member • Draft
13
DO NOT MERGE - pin vllm to main commit 9ea3a4015 for FP8 MoE+LoRA fix (vllm#42120)
#2849 opened Jun 21, 2026 by JohannesHa Member • Draft
fix(configs): validate inference weight_broadcast type against orchestrator
#2848 opened Jun 21, 2026 by EazyReal Contributor Loading…
feat: add r3 debug config
#2847 opened Jun 21, 2026 by mikasenghaas Member • Draft
feat(orchestrator): make target_lag a configurable field
#2845 opened Jun 20, 2026 by EazyReal Contributor Loading…
1
3
feat(orchestrator): use trace algorithm registry
#2842 opened Jun 20, 2026 by willccbb Member • Draft
11
Support v1 raw multimodal image offload
#2836 opened Jun 18, 2026 by eligotts Contributor • Draft
Halt after max_consecutive_errored_rollouts
#2835 opened Jun 18, 2026 by rasdani Contributor Loading…
9
Enable Ulysses CP for Qwen VLM training
#2834 opened Jun 18, 2026 by eligotts Contributor • Draft
6
Add linear length penalty as a summed GRPO advantage
#2829 opened Jun 16, 2026 by hallerite Member • Draft
Added softmax_scale in attentions
#2828 opened Jun 16, 2026 by Sirorezka Loading…
11
feat(configs): Kimi-K2.7-Code PD-disaggregated wide-EP inference config
#2827 opened Jun 15, 2026 by hallerite Member • Draft
chore: bump vLLM to 0.23.0 and prune/adapt monkeypatches
#2825 opened Jun 15, 2026 by hallerite Member Loading…
ProTip! Type g i on any issue or pull request to go back to the issue listing page.