[New Feature] Add the serving support #279
Open
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
cyber-pioneer
changed the title
ceci3
reviewed
Nov 27, 2024
| gpu-memory-utilization: 0.9 | ||
| max-model-len: 32768 | ||
| max-num-seqs: 256 | ||
| port: 4567 |
There was a problem hiding this comment.
是的 vllm字段内的参数，与原生的vllm serve命令行保持一致
| max-model-len: 32768 | ||
| max-num-seqs: 256 | ||
| port: 4567 | ||
| action-args: |
There was a problem hiding this comment.
举例说明：vllm serve /models/Qwen2.5-7B-Instruct --tensor-parallel-size=1 --gpu-memory-utilization=0.9 --max-model-len=32768 --max-num-seqs=256 --port=4567 --trust-remote-code --enable-chunked-prefill
--trust-remote-code --enable-chunked-prefill 这两个参数是预设的行为，命令行里是禁止传入参数值的，属于action args。
本PR将增加文档说明部署参数选项，降低开发者的接入成本。
| else: | ||
| run_local_command(f"bash {host_run_script_file}", dryrun) | ||
|
|
||
| def run(self, with_test=False, dryrun=False): |
| "localhost", | ||
| available_addr, | ||
| available_port, | ||
| 1, |
There was a problem hiding this comment.
本pr里仅支持单卡，多卡场景支持中
Comment on lines
+19
to
+20
| pip install modelscope | ||
| modelscope download --model Qwen/Qwen2.5-7B-Instruct --local_dir /models/ |
flagscale/serve/README.md
Outdated
| python run.py --config-path examples/qwen/ --config-name config action=run | ||
| ``` | ||
|
|
||
|
|
flagscale/serve/run_simple_vllm.py
Outdated
Comment on lines
34
to
40
| logger.info("Standard Output:") | ||
| logger.info(stdout) | ||
| # logger.info(stdout.decode()) | ||
| logger.info("Standard Error:") | ||
| logger.info(stderr) | ||
| # logger.info(stderr.decode()) | ||
|
|
cyber-pioneer
changed the title
aoyulong
changed the title
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.
New Feature
Flagscale supports serve.
Description
This pull request introduces support for deploying large models with FlagScale, leveraging the Ray framework for efficient orchestration and scalability. Currently, this implementation supports the Qwen model, enabling users to easily deploy and manage large-scale machine learning services.
Future Key features include:
More details in Serve.