phoenix-oss/llama-stack
1
0
Fork 0
forked from phoenix-oss/llama-stack-mirror
Code Pull requests Releases Packages 2 Activity Actions

Update embedding dimension singular

Browse source
This commit is contained in:
Ashwin Bharambe 2025-02-20 16:14:46 -08:00
parent 9436dd570d
commit 2608b6074f
9 changed files with 15 additions and 15 deletions
Download patch file Download diff file Expand all files Collapse all files

4
llama_stack/providers/remote/inference/fireworks/fireworks.py
View file

@ -237,8 +237,8 @@ class FireworksInferenceAdapter(ModelRegistryHelper, Inference, NeedsRequestProv
model = await self.model_store.get_model(model_id) model = await self.model_store.get_model(model_id)
kwargs = {} kwargs = {}
if model.metadata.get("embedding_dimensions"): if model.metadata.get("embedding_dimension"):
kwargs["dimensions"] = model.metadata.get("embedding_dimensions") kwargs["dimensions"] = model.metadata.get("embedding_dimension")
assert all(not content_has_media(content) for content in contents), ( assert all(not content_has_media(content) for content in contents), (
"Fireworks does not support media for embeddings" "Fireworks does not support media for embeddings"
) )

2
llama_stack/providers/remote/inference/fireworks/models.py
View file

@ -56,7 +56,7 @@ MODEL_ENTRIES = [
provider_model_id="nomic-ai/nomic-embed-text-v1.5", provider_model_id="nomic-ai/nomic-embed-text-v1.5",
model_type=ModelType.embedding, model_type=ModelType.embedding,
metadata={ metadata={
"embedding_dimensions": 768, "embedding_dimension": 768,
"context_length": 8192, "context_length": 8192,
}, },
), ),

4
llama_stack/providers/remote/inference/ollama/models.py
View file

@ -88,7 +88,7 @@ model_entries = [
aliases=["all-minilm"], aliases=["all-minilm"],
model_type=ModelType.embedding, model_type=ModelType.embedding,
metadata={ metadata={
"embedding_dimensions": 384, "embedding_dimension": 384,
"context_length": 512, "context_length": 512,
}, },
), ),
@ -96,7 +96,7 @@ model_entries = [
provider_model_id="nomic-embed-text", provider_model_id="nomic-embed-text",
model_type=ModelType.embedding, model_type=ModelType.embedding,
metadata={ metadata={
"embedding_dimensions": 768, "embedding_dimension": 768,
"context_length": 8192, "context_length": 8192,
}, },
), ),

4
llama_stack/providers/remote/inference/together/models.py
View file

@ -52,7 +52,7 @@ MODEL_ENTRIES = [
provider_model_id="togethercomputer/m2-bert-80M-8k-retrieval", provider_model_id="togethercomputer/m2-bert-80M-8k-retrieval",
model_type=ModelType.embedding, model_type=ModelType.embedding,
metadata={ metadata={
"embedding_dimensions": 768, "embedding_dimension": 768,
"context_length": 8192, "context_length": 8192,
}, },
), ),
@ -60,7 +60,7 @@ MODEL_ENTRIES = [
provider_model_id="togethercomputer/m2-bert-80M-32k-retrieval", provider_model_id="togethercomputer/m2-bert-80M-32k-retrieval",
model_type=ModelType.embedding, model_type=ModelType.embedding,
metadata={ metadata={
"embedding_dimensions": 768, "embedding_dimension": 768,
"context_length": 32768, "context_length": 32768,
}, },
), ),

4
llama_stack/providers/remote/inference/vllm/vllm.py
View file

@ -375,8 +375,8 @@ class VLLMInferenceAdapter(Inference, ModelsProtocolPrivate):
kwargs = {} kwargs = {}
assert model.model_type == ModelType.embedding assert model.model_type == ModelType.embedding
assert model.metadata.get("embedding_dimensions") assert model.metadata.get("embedding_dimension")
kwargs["dimensions"] = model.metadata.get("embedding_dimensions") kwargs["dimensions"] = model.metadata.get("embedding_dimension")
assert all(not content_has_media(content) for content in contents), "VLLM does not support media for embeddings" assert all(not content_has_media(content) for content in contents), "VLLM does not support media for embeddings"
response = self.client.embeddings.create( response = self.client.embeddings.create(
model=model.provider_resource_id, model=model.provider_resource_id,

2
llama_stack/templates/fireworks/run-with-safety.yaml
View file

@ -150,7 +150,7 @@ models:
provider_model_id: accounts/fireworks/models/llama-guard-3-11b-vision provider_model_id: accounts/fireworks/models/llama-guard-3-11b-vision
model_type: llm model_type: llm
- metadata: - metadata:
embedding_dimensions: 768 embedding_dimension: 768
context_length: 8192 context_length: 8192
model_id: nomic-ai/nomic-embed-text-v1.5 model_id: nomic-ai/nomic-embed-text-v1.5
provider_id: fireworks provider_id: fireworks

2
llama_stack/templates/fireworks/run.yaml
View file

@ -144,7 +144,7 @@ models:
provider_model_id: accounts/fireworks/models/llama-guard-3-11b-vision provider_model_id: accounts/fireworks/models/llama-guard-3-11b-vision
model_type: llm model_type: llm
- metadata: - metadata:
embedding_dimensions: 768 embedding_dimension: 768
context_length: 8192 context_length: 8192
model_id: nomic-ai/nomic-embed-text-v1.5 model_id: nomic-ai/nomic-embed-text-v1.5
provider_id: fireworks provider_id: fireworks

4
llama_stack/templates/together/run-with-safety.yaml
View file

@ -145,14 +145,14 @@ models:
provider_model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo provider_model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo
model_type: llm model_type: llm
- metadata: - metadata:
embedding_dimensions: 768 embedding_dimension: 768
context_length: 8192 context_length: 8192
model_id: togethercomputer/m2-bert-80M-8k-retrieval model_id: togethercomputer/m2-bert-80M-8k-retrieval
provider_id: together provider_id: together
provider_model_id: togethercomputer/m2-bert-80M-8k-retrieval provider_model_id: togethercomputer/m2-bert-80M-8k-retrieval
model_type: embedding model_type: embedding
- metadata: - metadata:
embedding_dimensions: 768 embedding_dimension: 768
context_length: 32768 context_length: 32768
model_id: togethercomputer/m2-bert-80M-32k-retrieval model_id: togethercomputer/m2-bert-80M-32k-retrieval
provider_id: together provider_id: together

4
llama_stack/templates/together/run.yaml
View file

@ -139,14 +139,14 @@ models:
provider_model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo provider_model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo
model_type: llm model_type: llm
- metadata: - metadata:
embedding_dimensions: 768 embedding_dimension: 768
context_length: 8192 context_length: 8192
model_id: togethercomputer/m2-bert-80M-8k-retrieval model_id: togethercomputer/m2-bert-80M-8k-retrieval
provider_id: together provider_id: together
provider_model_id: togethercomputer/m2-bert-80M-8k-retrieval provider_model_id: togethercomputer/m2-bert-80M-8k-retrieval
model_type: embedding model_type: embedding
- metadata: - metadata:
embedding_dimensions: 768 embedding_dimension: 768
context_length: 32768 context_length: 32768
model_id: togethercomputer/m2-bert-80M-32k-retrieval model_id: togethercomputer/m2-bert-80M-32k-retrieval
provider_id: together provider_id: together