feat: expose openai api endpoints from vllm #112
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
saghul
reviewed
Oct 24, 2024
80a3a80 to
4a4396b
Compare
saghul
reviewed
Oct 24, 2024
| vllm_server_path = os.environ.get('VLLM_SERVER_PATH', 'vllm.entrypoints.openai.api_server') | ||
| openai_api_server_port = int(os.environ.get('OPENAI_API_SERVER_PORT', 8003)) | ||
| openai_api_base_url = os.environ.get('OPENAI_API_BASE_URL', f'http://localhost:{openai_api_server_port}') | ||
| openai_api_server_port = int(os.environ.get('OPENAI_API_SERVER_PORT', app_port if use_vllm else 8003)) |
There was a problem hiding this comment.
Shall we get rid of this while we're here, and use the URL, defaulting to ollama's default port when not doing vllm?
saghul
reviewed
Oct 24, 2024
| sys.exit(1) | ||
|
|
||
| log.info(f'Enabled modules: {modules}') | ||
|
|
||
| if device == 'cuda' or is_mac: | ||
| log.info('Using GPU') |
saghul
approved these changes
Oct 24, 2024
added 2 commits
October 24, 2024 14:08
shooding
added a commit
to shooding/skynet
that referenced
this pull request
Nov 26, 2024
* master: (61 commits) deps: use pypi provided silero vad, upgrade to latest fix: remove public key validation (jitsi#123) fix: downgrade vllm (jitsi#122) feat: add fallback folder when looking up public keys (jitsi#119) fix: add ffmpeg dependency for pytorch ref: bypass queueing jobs with invalid payload (jitsi#121) fix: replace examplar usage with label for app_id feat: add instrumentation for app_id (jitsi#118) fix: re-enable vLLM multiprocessing (jitsi#116) fix: update incorrect prompt example fix: healthchecks failing due to missing internal id (jitsi#115) feat(openai-api) use Ollama for local development feat: expose openai api endpoints from vllm (jitsi#112) feat: update text hint type prompting (jitsi#111) feat: add meeting hint type and use it as default (jitsi#110) feat: enable requests batching (jitsi#109) metrics: add full duration metric metrics: add a skipped job status which will not count towards duration metrics fix: catch exceptions when echoing fails feat: add support for echoing requests (jitsi#107) ... # Conflicts: # Dockerfile # Makefile # requirements.txt
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.
No description provided.