[Perf] Alexsander fixes round 3 - Oct 25th #15935
Open
Conversation
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
Replace json.dumps with orjson.dumps in HTTP handler to reduce serialization latency for all LLM provider API calls.
Replace json.dumps/loads with orjson in streaming hot paths. Saves ~350ms per 1000-chunk streaming response.
- Move orjson from optional to required dependencies in pyproject.toml - Fixes ModuleNotFoundError when importing litellm core modules - orjson is used in llm_http_handler.py which is imported by core litellm
- Optimize jsonify_object() (24 call sites across codebase) - Optimize get_request_status() for metadata parsing Reduces CPU usage and improves database write latency.
- serialize_object() uses orjson for 3-5x faster dict serialization - get_prompt_caching_cache_key() skips encode/decode cycle - Maintains cache key compatibility
- Maintains circular reference detection - Preserves default=str fallback behavior - Used in 15+ files across proxy and integrations
- Added @lru_cache(1024) to get_cooldown_cache_key() - Changed all 4 locations to use the cached method instead of recreating strings - Replaced f-strings with string concatenation for better performance Results: get_cooldown_cache_key dropped from 47MB to a few bytes and is no longer a top consumer. Memory leak still present. Next: Optimize heavy memory consumers so memory leaks become more obvious.
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
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Outdated suggestions cannot be applied.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.
Title
[Perf] Alexsander fixes round 3 - Oct 25th
Relevant issues
Pre-Submission checklist
Please complete all items before asking a LiteLLM maintainer to review your PR
tests/litellm/directory, Adding at least 1 test is a hard requirement - see detailsmake test-unitType
🐛 Bug Fix
🧹 Refactoring
Changes