[pull] main from inclusionAI:main#29
Merged
pull[bot] merged 7 commits intoaxistore80-coder:mainfrom Apr 8, 2026
Merged
Conversation
Keep no-drop evaluation stable when batches do not divide evenly across DP workers. Pad eval inputs with zero-weight dummy items and make engine loss reduction tolerate local empty shards so distributed and pipeline-parallel paths stay synchronized. Key changes: - pad evaluate_* dispatches to preserve DP and RW pairing invariants - skip zero-weight local loss work in FSDP, Megatron, and Archon - add eval dispatch regression tests and refresh CLI reference docs
* feat(ci): separate vllm and sglang pyproject.toml * fix(ci): support vllm pyproject in docker and install tests * fix(ci): correct Dockerfile RUN chaining for pyproject swap * fix(ci): avoid read-only pyproject bind mount writes * fix(ci): validate docker variant and sync vllm lockfile docs Fail fast on invalid Docker VARIANT values to prevent silently building the wrong backend image, and align vLLM setup instructions with CI by copying uv.vllm.lock when swapping pyproject variants. Key changes: - validate VARIANT via case statement in Dockerfile - update README/docs/agent guidance with uv.vllm.lock copy step - regenerate CLI reference docs via pre-commit hook
…rging PP shards (#1145) * fix(engine): XCCL lora weights update was being overwritten when pp>1 * chore: addressed gemini comments
Ranks with no gradients (e.g. frozen non-LoRA params) previously returned 0.0 immediately, skipping the all_reduce. Ranks that do have gradients then hang waiting for the collective to complete. Move device init before the empty-grads check and make zero-grad ranks still participate in all_reduce with a zero-valued tensor. Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
See Commits and Changes for more details.
Created by
pull[bot] (v2.0.0-alpha.4)
Can you help keep this open source service alive? 💖 Please sponsor : )