-
Notifications
You must be signed in to change notification settings - Fork 308
Pull requests: PrimeIntellect-ai/prime-rl
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix: is_tt_moe_model checks text_config for VLM-style MoE models
#2780
opened Jun 11, 2026 by
Levichev
Loading…
feat: add RL checkpoint format backward compat integration test
#2776
opened Jun 11, 2026 by
samsja
Member
Loading…
feat(orchestrator): add penalize action for gibberish/repetition filters
#2775
opened Jun 11, 2026 by
anravich13-cloud
Loading…
fix(inference): patch vLLM 0.22 O(B*L) sampler hot paths
#2772
opened Jun 11, 2026 by
joanvelja
Loading…
feat: consume native multimodal from the v1 trace (v0 + v1 VLM training)
#2751
opened Jun 10, 2026 by
mikasenghaas
Member
•
Draft
feat: algorithm abstraction — named algorithm classes + inline frozen-model references (grpo, opd, sft_distill, self_distill, echo)
#2746
opened Jun 9, 2026 by
hallerite
Member
Loading…
feat(orchestrator): EnvMixStrategy seam for env selection
#2743
opened Jun 9, 2026 by
hallerite
Member
Loading…
feat: dynamo inference backend integration
#2737
opened Jun 9, 2026 by
biswapanda
Loading…
1 task done
chore: bump research-environments + verifiers, add swebench-pro
#2719
opened Jun 4, 2026 by
mikasenghaas
Member
•
Draft
test: add renderer client tests for chat_template_kwargs materialization
#2711
opened Jun 4, 2026 by
mvanhorn
Loading…
fix: guard fp32 lm-head logits to contiguous to avoid vLLM NaN
#2710
opened Jun 4, 2026 by
mvanhorn
Loading…
Previous Next
ProTip!
Follow long discussions with comments:>50.