feat:support for two long-context embedding models (Qwen3 and Gemma) #453
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
….6B and EmbeddingGemma-300M) Signed-off-by: OneZero-Y <[email protected]> feat:support for two long-context embedding models (Qwen3-Embedding-0.6B and EmbeddingGemma-300M) Signed-off-by: OneZero-Y <[email protected]>
OneZero-Y
requested review from
Xunzhuo,
rootfs and
wangchen615
as code owners
👥 vLLM Semantic Team NotificationThe following members have been identified for the changed files in this PR and have been automatically assigned: 📁
|
|
Test failure is fixed in main branch, merging it for testing. |
rootfs
merged commit fa4f5c7
into
vllm-project:feat-candle-refactoring
3 of 4 checks passed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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Outdated suggestions cannot be applied.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.
What type of PR is this?
support for two long-context embedding models (Qwen3-Embedding-0.6B and EmbeddingGemma-300M)
What this PR does / why we need it:
1. Qwen3-Embedding-0.6B
Model Specifications:
Key Features:
2. EmbeddingGemma-300M
Model Specifications:
Key Features:
3. Intelligent Routing System
Routing Logic:
quality_priority > 0.7latency_priority > 0.7latency ≤ 0.7)512 < seq_len ≤ 4096+ Balancedseq_len > 4096Features:
4. Enhanced API Endpoints
4.1 Embedding Generation API
Endpoint:
POST /api/v1/embeddingsRequest:
Response:
4.2 Cosine Similarity Calculation API
Endpoint:
POST /api/v1/similarityRequest:
Response:
{"model_used":"qwen3","similarity":0.74183154,"processing_time_ms":30.0131}4.3 Batch Similarity Matching API
Endpoint:
POST /api/v1/similarity/batchRequest:
Response:
{"matches":[{"index":2,"similarity":0.9605684,"text":"deep learning"},{"index":0,"similarity":0.9055326,"text":"artificial intelligence"}],"total_candidates":4,"model_used":"gemma","processing_time_ms":356.3315}4.4 Embedding Models Information API
Endpoint:
GET /api/v1/embeddings/modelsRequest:
Response:
{"count":2,"models":[{"name":"qwen3_embedding_model","type":"embedding","loaded":true,"model_path":"models/Qwen3-Embedding-0.6B","metadata":{"default_dimension":"1024","matryoshka_supported":"true","max_sequence_length":"32768","model_type":"qwen3"}},{"name":"gemma_embedding_model","type":"embedding","loaded":true,"model_path":"models/embeddinggemma-300m","metadata":{"default_dimension":"768","matryoshka_supported":"true","max_sequence_length":"8192","model_type":"gemma"}}]}Configuration
config.yaml
Which issue(s) this PR fixes:
part of #266
Release Notes: Yes/No