feat: Implement proper sparse attention in API server#2
Open
AlexCuadron wants to merge 4 commits intomainfrom
Open
feat: Implement proper sparse attention in API server#2AlexCuadron wants to merge 4 commits intomainfrom
AlexCuadron wants to merge 4 commits intomainfrom
Conversation
- Add detailed client guide with examples in multiple languages - Add performance considerations and best practices - Add detailed configuration documentation - Add error handling documentation
- Replace basic model with proper sparse attention implementation - Add support for different model architectures (LLaMA, Mistral) - Add channel configuration support - Update API to be fully OpenAI-compatible - Add proper token streaming implementation - Add configuration for sparse attention parameters
- Add Qwen2 sparse attention implementation - Update configuration to support Qwen2 - Update documentation with Qwen2 support - Improve architecture selection documentation
- Create single server script that matches perplexity_eval.py usage - Automatic architecture detection from model config - Same command-line arguments as perplexity_eval.py - Remove need for manual architecture configuration
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR updates the API server to properly implement sparse attention from the core DoubleSparse implementation.
Changes
Core Changes
Configuration
MODEL_ARCHITECTUREsetting for model type selectionCHANNELsetting for channel selection (q, k, qk)Implementation Details