Detokenize incrementally when streaming #653
Merged
Conversation
This file contains 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
|
@hnyls2002 Nice work! cc @merrymercy @Ying1123 @yzh119 model: Llama 3 8B Instruct # server
python3 -m sglang.launch_server --model /root/Meta-Llama-3-8B-Instruct --trust-remote-code --port 23333 --disable-radix-cach --stream-interval 1
# client
python3 benchmark_serving.py --backend openai --host 127.0.0.1 --port 23333 --dataset /root/ShareGPT_V3_unfiltered_cleaned_split.json --model /root/Meta-Llama-3-8B-Instruct --tokenizer /root/Meta-Llama-3-8B-Instruct --num-prompts 1000 --request-rate 128 --trust-remote-code |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
The
detokenizer_managerachieves very low performance when 1) streaming is enabled, 2) model forward is very swift.Changed the decoding process into incremental detokenization in detokenizer_manager, and it is not compatible with mutable output_ids (when jump forward).