docs: add 'Using SGLang as the Inference Backend' guide (en/zh) - #5047
Open
JustinTong0323 wants to merge 2 commits into
Open
docs: add 'Using SGLang as the Inference Backend' guide (en/zh)#5047JustinTong0323 wants to merge 2 commits into
JustinTong0323 wants to merge 2 commits into
Conversation
MinerU2.5 is a Qwen2-VL-based model that SGLang serves natively. Document running SGLang as an OpenAI-compatible server and connecting MinerU's existing vlm-http-client backend to it (no code changes, no local torch on the client). Adds en + zh pages and registers them in the mkdocs nav.
dosubot
Bot
added
size:L
Contributor
|
I have read the CLA Document and I hereby sign the CLA JustinTong0323 seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account. |
Measured ~10% higher end-to-end throughput on the OmniDocBench two-step workload (single H200: ~153 vs ~135 pages/min) by passing image tensors over CUDA IPC; output unchanged. Added to both the en and zh guides.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
Adds a documentation guide "Using SGLang as the Inference Backend" (English + Simplified Chinese) and registers it in the mkdocs nav.
MinerU2.5 is a
Qwen2-VL-based model (Qwen2VLForConditionalGeneration) that SGLang serves natively. MinerU shipped a nativevlm-sglang-enginefor older releases and switched its default acceleration backend to vLLM for 2.5 — but MinerU's genericvlm-http-clientbackend talks to any OpenAI-compatible server, including SGLang. This guide documents that path (no code changes, no localtorchon the client).Contents
--chat-template/--trust-remote-codeare unnecessary;sglang servealias shown).mineru -p <in> -o <out> -b vlm-http-client -u http://127.0.0.1:30000, plus themineru-vl-utilsPython path.no_repeat_ngram_size; mitigations +--enable-custom-logit-processor).Verified end-to-end on SGLang 0.5.12.post1 (full OmniDocBench v1.6 run, 1651 pages).