feat: add nemo retriever text embedding models to nvidia inference provider (#1218)

# What does this PR do?

add the NeMo Retriever Embedding models from
https://docs.nvidia.com/nim/nemo-retriever/text-embedding/latest/support-matrix.html

docs/source/distributions/remote_hosted_distro/nvidia.md

@@ -36,7 +36,10 @@ The following models are available by default:
 - `meta-llama/Llama-3.2-3B-Instruct (meta/llama-3.2-3b-instruct)`
 - `meta-llama/Llama-3.2-11B-Vision-Instruct (meta/llama-3.2-11b-vision-instruct)`
 - `meta-llama/Llama-3.2-90B-Vision-Instruct (meta/llama-3.2-90b-vision-instruct)`
-- `baai/bge-m3 (baai/bge-m3)`
+- `nvidia/llama-3.2-nv-embedqa-1b-v2 (nvidia/llama-3.2-nv-embedqa-1b-v2)`
+- `nvidia/nv-embedqa-e5-v5 (nvidia/nv-embedqa-e5-v5)`
+- `nvidia/nv-embedqa-mistral-7b-v2 (nvidia/nv-embedqa-mistral-7b-v2)`
+- `snowflake/arctic-embed-l (snowflake/arctic-embed-l)`
 
 
 ### Prerequisite: API Keys

llama_stack/providers/remote/inference/nvidia/models.py

@@ -48,12 +48,49 @@ _MODEL_ENTRIES = [
         "meta/llama-3.2-90b-vision-instruct",
         CoreModelId.llama3_2_90b_vision_instruct.value,
     ),
+    # NeMo Retriever Text Embedding models -
+    #
+    # https://docs.nvidia.com/nim/nemo-retriever/text-embedding/latest/support-matrix.html
+    #
+    # +-----------------------------------+--------+-----------+-----------+------------+
+    # | Model ID                          | Max    | Publisher | Embedding | Dynamic    |
+    # |                                   | Tokens |           | Dimension | Embeddings |
+    # +-----------------------------------+--------+-----------+-----------+------------+
+    # | nvidia/llama-3.2-nv-embedqa-1b-v2 | 8192   | NVIDIA    | 2048      | Yes        |
+    # | nvidia/nv-embedqa-e5-v5           |  512   | NVIDIA    | 1024      |  No        |
+    # | nvidia/nv-embedqa-mistral-7b-v2   |  512   | NVIDIA    | 4096      |  No        |
+    # | snowflake/arctic-embed-l          |  512   | Snowflake | 1024      |  No        |
+    # +-----------------------------------+--------+-----------+-----------+------------+
     ProviderModelEntry(
-        provider_model_id="baai/bge-m3",
+        provider_model_id="nvidia/llama-3.2-nv-embedqa-1b-v2",
+        model_type=ModelType.embedding,
+        metadata={
+            "embedding_dimension": 2048,
+            "context_length": 8192,
+        },
+    ),
+    ProviderModelEntry(
+        provider_model_id="nvidia/nv-embedqa-e5-v5",
         model_type=ModelType.embedding,
         metadata={
             "embedding_dimension": 1024,
-            "context_length": 8192,
+            "context_length": 512,
+        },
+    ),
+    ProviderModelEntry(
+        provider_model_id="nvidia/nv-embedqa-mistral-7b-v2",
+        model_type=ModelType.embedding,
+        metadata={
+            "embedding_dimension": 4096,
+            "context_length": 512,
+        },
+    ),
+    ProviderModelEntry(
+        provider_model_id="snowflake/arctic-embed-l",
+        model_type=ModelType.embedding,
+        metadata={
+            "embedding_dimension": 1024,
+            "context_length": 512,
         },
     ),
     # TODO(mf): how do we handle Nemotron models?

llama_stack/templates/nvidia/run.yaml

@@ -136,11 +136,32 @@ models:
   provider_model_id: meta/llama-3.2-90b-vision-instruct
   model_type: llm
 - metadata:
-    embedding_dimension: 1024
+    embedding_dimension: 2048
     context_length: 8192
-  model_id: baai/bge-m3
+  model_id: nvidia/llama-3.2-nv-embedqa-1b-v2
   provider_id: nvidia
-  provider_model_id: baai/bge-m3
+  provider_model_id: nvidia/llama-3.2-nv-embedqa-1b-v2
+  model_type: embedding
+- metadata:
+    embedding_dimension: 1024
+    context_length: 512
+  model_id: nvidia/nv-embedqa-e5-v5
+  provider_id: nvidia
+  provider_model_id: nvidia/nv-embedqa-e5-v5
+  model_type: embedding
+- metadata:
+    embedding_dimension: 4096
+    context_length: 512
+  model_id: nvidia/nv-embedqa-mistral-7b-v2
+  provider_id: nvidia
+  provider_model_id: nvidia/nv-embedqa-mistral-7b-v2
+  model_type: embedding
+- metadata:
+    embedding_dimension: 1024
+    context_length: 512
+  model_id: snowflake/arctic-embed-l
+  provider_id: nvidia
+  provider_model_id: snowflake/arctic-embed-l
   model_type: embedding
 shields: []
 vector_dbs: []