Document benchmarking for PD deployment with GLM-5.1 - #58
Open
yiminghub2024 wants to merge 3 commits into
Open
Conversation
…ts own chunk (tile-ai#55) Two fixes for serving OpenAI-compatible clients over a PD deployment. The prefill request now drops `stream_options` and `max_completion_tokens` from the client body: the first contradicts the `stream=False` we force (vLLM rejects the pair with a 400 during body parsing) and the second overrides our `max_tokens=1`. Verified with `vllm bench serve --backend openai-chat`, which sends both unconditionally and previously failed every request. `ignore_eos` is now forwarded to the decode engine and honoured by the MLA/NSA adapter, streaming usage moves into its own trailing chunk carrying `total_tokens` (the shape vLLM and the OpenAI API emit), and the `transformers` / `tokenizers` pins are relaxed to `>=`. --------- Co-authored-by: CrimsonDump <56749892+CrimsonDump@users.noreply.github.com> Co-authored-by: Shuang Wu <leo@aigcsemi.com>
Update the README for the v0.1.5.post2 release: install commands and expected version now reference `0.1.5.post2`, the News and wheel download links point to the v0.1.5.post2 release (the previous `v0.1.5` release URL does not exist), and the `transformers` / `tokenizers` rows reflect the relaxed `>=` pins. Co-authored-by: Ziming Miao <max@aigcsemi.com>
Added benchmarking details for PD deployment using GLM-5.1 and vllm with TileRT. Included latency and throughput metrics for performance evaluation.
Author
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 join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Added benchmarking details for PD deployment using GLM-5.1 and vllm with TileRT. Included latency and throughput metrics for performance evaluation.