cerebras template update for memory (#792)

# What does this PR do?

- we no longer have meta-reference as memory provider, update cerebras
template


## Test Plan

```
python llama_stack/scripts/distro_codegen.py
```

## Sources

Please link relevant resources if necessary.


## Before submitting

- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Ran pre-commit to handle lint / formatting issues.
- [ ] Read the [contributor
guideline](https://github.com/meta-llama/llama-stack/blob/main/CONTRIBUTING.md),
      Pull Request section?
- [ ] Updated relevant documentation.
- [ ] Wrote necessary unit or integration tests.

distributions/dependencies.json

@@ -346,6 +346,7 @@
     "blobfile",
     "cerebras_cloud_sdk",
     "chardet",
+    "chromadb-client",
     "datasets",
     "faiss-cpu",
     "fastapi",

docs/source/distributions/self_hosted_distro/cerebras.md

@@ -8,7 +8,7 @@ The `llamastack/distribution-cerebras` distribution consists of the following pr
 | datasetio | `remote::huggingface`, `inline::localfs` |
 | eval | `inline::meta-reference` |
 | inference | `remote::cerebras` |
-| memory | `inline::meta-reference` |
+| memory | `inline::faiss`, `remote::chromadb`, `remote::pgvector` |
 | safety | `inline::llama-guard` |
 | scoring | `inline::basic`, `inline::llm-as-judge`, `inline::braintrust` |
 | telemetry | `inline::meta-reference` |

llama_stack/templates/bedrock/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: bedrock
 distribution_spec:
   description: Use AWS Bedrock for running LLM inference and safety
   providers:

llama_stack/templates/bedrock/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: bedrock
-conda_env: bedrock
 apis:
 - agents
 - datasetio

llama_stack/templates/cerebras/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: cerebras
 distribution_spec:
   description: Use Cerebras for running LLM inference
   providers:
@@ -8,7 +7,9 @@ distribution_spec:
     safety:
     - inline::llama-guard
     memory:
-    - inline::meta-reference
+    - inline::faiss
+    - remote::chromadb
+    - remote::pgvector
     agents:
     - inline::meta-reference
     eval:

llama_stack/templates/cerebras/cerebras.py

@@ -27,7 +27,7 @@ def get_distribution_template() -> DistributionTemplate:
     providers = {
         "inference": ["remote::cerebras"],
         "safety": ["inline::llama-guard"],
-        "memory": ["inline::meta-reference"],
+        "memory": ["inline::faiss", "remote::chromadb", "remote::pgvector"],
         "agents": ["inline::meta-reference"],
         "eval": ["inline::meta-reference"],
         "datasetio": ["remote::huggingface", "inline::localfs"],

llama_stack/templates/cerebras/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: cerebras
-conda_env: cerebras
 apis:
 - agents
 - datasetio
@@ -26,13 +25,19 @@ providers:
     provider_type: inline::llama-guard
     config: {}
   memory:
-  - provider_id: meta-reference
+  - provider_id: faiss
-    provider_type: inline::meta-reference
+    provider_type: inline::faiss
     config:
       kvstore:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/cerebras}/faiss_store.db
+  - provider_id: chromadb
+    provider_type: remote::chromadb
+    config: {}
+  - provider_id: pgvector
+    provider_type: remote::pgvector
+    config: {}
   agents:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/fireworks/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: fireworks
 distribution_spec:
   description: Use Fireworks.AI for running LLM inference
   providers:

llama_stack/templates/fireworks/run-with-safety.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: fireworks
-conda_env: fireworks
 apis:
 - agents
 - datasetio

llama_stack/templates/fireworks/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: fireworks
-conda_env: fireworks
 apis:
 - agents
 - datasetio

llama_stack/templates/hf-endpoint/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: hf-endpoint
 distribution_spec:
   description: Use (an external) Hugging Face Inference Endpoint for running LLM inference
   providers:

llama_stack/templates/hf-endpoint/run-with-safety.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: hf-endpoint
-conda_env: hf-endpoint
 apis:
 - agents
 - datasetio

llama_stack/templates/hf-endpoint/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: hf-endpoint
-conda_env: hf-endpoint
 apis:
 - agents
 - datasetio

llama_stack/templates/hf-serverless/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: hf-serverless
 distribution_spec:
   description: Use (an external) Hugging Face Inference Endpoint for running LLM inference
   providers:

llama_stack/templates/hf-serverless/run-with-safety.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: hf-serverless
-conda_env: hf-serverless
 apis:
 - agents
 - datasetio

llama_stack/templates/hf-serverless/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: hf-serverless
-conda_env: hf-serverless
 apis:
 - agents
 - datasetio

llama_stack/templates/meta-reference-gpu/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: meta-reference-gpu
 distribution_spec:
   description: Use Meta Reference for running LLM inference
   providers:

llama_stack/templates/meta-reference-gpu/run-with-safety.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: meta-reference-gpu
-conda_env: meta-reference-gpu
 apis:
 - agents
 - datasetio

llama_stack/templates/meta-reference-gpu/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: meta-reference-gpu
-conda_env: meta-reference-gpu
 apis:
 - agents
 - datasetio

llama_stack/templates/meta-reference-quantized-gpu/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: meta-reference-quantized-gpu
 distribution_spec:
   description: Use Meta Reference with fp8, int4 quantization for running LLM inference
   providers:

llama_stack/templates/meta-reference-quantized-gpu/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: meta-reference-quantized-gpu
-conda_env: meta-reference-quantized-gpu
 apis:
 - agents
 - datasetio

llama_stack/templates/nvidia/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: nvidia
 distribution_spec:
   description: Use NVIDIA NIM for running LLM inference
   providers:

llama_stack/templates/nvidia/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: nvidia
-conda_env: nvidia
 apis:
 - agents
 - datasetio

llama_stack/templates/ollama/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: ollama
 distribution_spec:
   description: Use (an external) Ollama server for running LLM inference
   providers:

llama_stack/templates/ollama/run-with-safety.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: ollama
-conda_env: ollama
 apis:
 - agents
 - datasetio

llama_stack/templates/ollama/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: ollama
-conda_env: ollama
 apis:
 - agents
 - datasetio

llama_stack/templates/remote-vllm/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: remote-vllm
 distribution_spec:
   description: Use (an external) vLLM server for running LLM inference
   providers:

llama_stack/templates/remote-vllm/run-with-safety.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: remote-vllm
-conda_env: remote-vllm
 apis:
 - agents
 - inference

llama_stack/templates/remote-vllm/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: remote-vllm
-conda_env: remote-vllm
 apis:
 - agents
 - inference

llama_stack/templates/tgi/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: tgi
 distribution_spec:
   description: Use (an external) TGI server for running LLM inference
   providers:

llama_stack/templates/tgi/run-with-safety.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: tgi
-conda_env: tgi
 apis:
 - agents
 - datasetio

llama_stack/templates/tgi/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: tgi
-conda_env: tgi
 apis:
 - agents
 - datasetio

llama_stack/templates/together/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: together
 distribution_spec:
   description: Use Together.AI for running LLM inference
   providers:

llama_stack/templates/together/run-with-safety.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: together
-conda_env: together
 apis:
 - agents
 - datasetio

llama_stack/templates/together/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: together
-conda_env: together
 apis:
 - agents
 - datasetio

llama_stack/templates/vllm-gpu/build.yaml

@@ -1,5 +1,4 @@
 version: '2'
-name: vllm-gpu
 distribution_spec:
   description: Use a built-in vLLM engine for running LLM inference
   providers:

llama_stack/templates/vllm-gpu/run.yaml

@@ -1,6 +1,5 @@
 version: '2'
 image_name: vllm-gpu
-conda_env: vllm-gpu
 apis:
 - agents
 - datasetio