feat: small ollama package

llama_stack/templates/ollama/build.yaml

@@ -25,23 +25,7 @@ distribution_spec:
     - inline::rag-runtime
     - remote::model-context-protocol
     - remote::wolfram-alpha
-  metadata_store:
-    type: sqlite
-    db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ollama}/registry.db
-  inference_store:
-    type: sqlite
-    db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ollama}/inference_store.db
-  models:
-  - metadata: {}
-    model_id: ${env.INFERENCE_MODEL}
-    provider_id: remote::ollama
-    model_type: llm
-  - metadata:
-      embedding_dimension: 384
-    model_id: all-MiniLM-L6-v2
-    provider_id: remote::ollama
-    provider_model_id: all-minilm:latest
-    model_type: embedding
-image_type: container
+image_type: conda
 additional_pip_packages:
 - sqlalchemy[asyncio]
+- blobfile

llama_stack/templates/ollama/ollama.py

@@ -13,8 +13,8 @@ from llama_stack.distribution.datatypes import (
     ShieldInput,
     ToolGroupInput,
 )
-from llama_stack.providers.inline.post_training.huggingface import HuggingFacePostTrainingConfig
-from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
+#from llama_stack.providers.inline.post_training.huggingface import HuggingFacePostTrainingConfig
+#from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
 from llama_stack.providers.remote.inference.ollama import OllamaImplConfig
 from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
 
@@ -32,7 +32,6 @@ def get_distribution_template() -> DistributionTemplate:
         "tool_runtime": [
             "remote::brave-search",
             "remote::tavily-search",
-            "inline::rag-runtime",
             "remote::model-context-protocol",
             "remote::wolfram-alpha",
         ],
@@ -43,11 +42,11 @@ def get_distribution_template() -> DistributionTemplate:
         provider_type="remote::ollama",
         config=OllamaImplConfig.sample_run_config(),
     )
-    vector_io_provider_faiss = Provider(
-        provider_id="faiss",
-        provider_type="inline::faiss",
-        config=FaissVectorIOConfig.sample_run_config(f"~/.llama/distributions/{name}"),
-    )
+    #vector_io_provider_faiss = Provider(
+    #    provider_id="faiss",
+    #    provider_type="inline::faiss",
+    #    config=FaissVectorIOConfig.sample_run_config(f"~/.llama/distributions/{name}"),
+    #)
     inference_model = ModelInput(
         model_id="${env.INFERENCE_MODEL}",
         provider_id="ollama",
@@ -70,10 +69,6 @@ def get_distribution_template() -> DistributionTemplate:
             toolgroup_id="builtin::websearch",
             provider_id="tavily-search",
         ),
-        ToolGroupInput(
-            toolgroup_id="builtin::rag",
-            provider_id="rag-runtime",
-        ),
         ToolGroupInput(
             toolgroup_id="builtin::wolfram_alpha",
             provider_id="wolfram-alpha",

llama_stack/templates/ollama/run.yaml

@@ -24,6 +24,10 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ollama}/faiss_store.db
+  - provider_id: chromadb
+    provider_type: remote::chromadb
+    config:
+      url: ${env.CHROMADB_URL:http://host.docker.internal:8000}
   safety:
   - provider_id: llama-guard
     provider_type: inline::llama-guard