10 commits

Author	SHA1	Message	Date
Francisco Javier Arceo	62005dc1a9	feat: Making static prompt values in Rag/File Search configurable in Vector Store Config (#4368) ... # What does this PR do? - Enables users to configure prompts used throughout the File Search / Vector Retrieval - Configuration is defined in the Vector Stores Config so they can be modified at runtime - Backwards compatible, which means the fields are optional and default to the previously used values This is the summary of the new options in the `run.yaml` ```yaml vector_stores: file_search_params: header_template: 'knowledge_search tool found {num_chunks} chunks:\nBEGIN of knowledge_search tool results.\n' footer_template: 'END of knowledge_search tool results.\n' context_prompt_params: chunk_annotation_template: 'Result {index}\nContent: {chunk.content}\nMetadata: {metadata}\n' context_template: 'The above results were retrieved to help answer the user\'s query: "{query}". Use them as supporting information only in answering this query.{annotation_instruction}\n' annotation_prompt_params: enable_annotations: true annotation_instruction_template: 'Cite sources immediately at the end of sentences before punctuation, using `<|file-id|>` format like \'This is a fact <|file-Cn3MSNn72ENTiiq11Qda4A|>.\'. Do not add extra punctuation. Use only the file IDs provided, do not invent new ones.' chunk_annotation_template: '[{index}] {metadata_text} cite as <|{file_id}|>\n{chunk_text}\n' ``` <!-- If resolving an issue, uncomment and update the line below --> <!-- Closes #[issue-number] --> ## Test Plan Added tests. --------- Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>	2025-12-15 11:39:01 -05:00
Jaideep Rao	76e47d811a	feat(api): add readonly connectors API (#4258) ... # What does this PR do? Adds a new API for connectors and MCP registry support along with required types. Does not include any implementation for it <!-- If resolving an issue, uncomment and update the line below --> Closes #4235 and #4061 (partially) ## Test Plan no tests included --------- Signed-off-by: Jaideep Rao <jrao@redhat.com> Co-authored-by: Francisco Javier Arceo <arceofrancisco@gmail.com>	2025-12-11 10:19:55 -08:00
Francisco Javier Arceo	95b2948d11	feat: Add support for query rewrite in vector_store.search (#4171) ... # What does this PR do? Actualize query rewrite in search API, add `default_query_expansion_model` and `query_expansion_prompt` in `VectorStoresConfig`. Makes `rewrite_query` parameter functional in vector store search. - `rewrite_query=false` (default): Use original query - `rewrite_query=true`: Expand query via LLM, or fail gracefully if no LLM available Adds 4 parameters to`VectorStoresConfig`: - `default_query_expansion_model`: LLM model for query expansion (optional) - `query_expansion_prompt`: Custom prompt template (optional, uses built-in default) - `query_expansion_max_tokens`: Configurable token limit (default: 100) - `query_expansion_temperature`: Configurable temperature (default: 0.3) Enabled `run.yaml`: ```yaml vector_stores: rewrite_query_params: model: provider_id: "ollama" model_id: "llama3.2:3b-instruct-fp16" # prompt defaults to built-in # max_tokens defaults to 100 # temperature defaults to 0.3 ``` Fully customized `run.yaml`: ```yaml vector_stores: default_provider_id: faiss default_embedding_model: provider_id: sentence-transformers model_id: nomic-ai/nomic-embed-text-v1.5 rewrite_query_params: model: provider_id: ollama model_id: llama3.2:3b-instruct-fp16 prompt: "Rewrite this search query to improve retrieval results by expanding it with relevant synonyms and related terms: {query}" max_tokens: 100 temperature: 0.3 ``` ## Test Plan Added test and recording Example script as well: ```python import asyncio from llama_stack_client import LlamaStackClient from io import BytesIO def gen_file(client, text: str=""): file_buffer = BytesIO(text.encode('utf-8')) file_buffer.name = "my_file.txt" uploaded_file = client.files.create( file=file_buffer, purpose="assistants" ) return uploaded_file async def test_query_rewriting(): client = LlamaStackClient(base_url="http://0.0.0.0:8321/") uploaded_file = gen_file(client, "banana banana apple") uploaded_file2 = gen_file(client, "orange orange kiwi") vs = client.vector_stores.create() xf_vs = client.vector_stores.files.create(vector_store_id=vs.id, file_id=uploaded_file.id) xf_vs1 = client.vector_stores.files.create(vector_store_id=vs.id, file_id=uploaded_file2.id) response1 = client.vector_stores.search( vector_store_id=vs.id, query="apple", max_num_results=3, rewrite_query=False ) response2 = client.vector_stores.search( vector_store_id=vs.id, query="kiwi", max_num_results=3, rewrite_query=True, ) print(f"\n🔵 Response 1 (rewrite_query=False):\n\033[94m{response1}\033[0m") print(f"\n🟢 Response 2 (rewrite_query=True):\n\033[92m{response2}\033[0m") for f in [uploaded_file.id, uploaded_file2.id]: client.files.delete(file_id=f) client.vector_stores.delete(vector_store_id=vs.id) if __name__ == "__main__": asyncio.run(test_query_rewriting()) ``` And see the screen shot of the server logs showing it worked. <img width="1111" height="826" alt="Screenshot 2025-11-19 at 1 16 03 PM" src="https://github.com/user-attachments/assets/2d188b44-1fef-4df5-b465-2d6728ca49ce" /> Notice the log: ```bash Query rewritten: 'kiwi' → 'kiwi, a small brown or green fruit native to New Zealand, or a person having a fuzzy brown outer skin similar in appearance.' ``` So `kiwi` was expanded. --------- Signed-off-by: Francisco Javier Arceo <farceo@redhat.com> Co-authored-by: Matthew Farrellee <matt@cs.wisc.edu>	2025-12-10 10:06:19 -05:00
Charlie Doern	661985e240	feat: remove usage of build yaml (#4192) ... # What does this PR do? the build.yaml is only used in the following ways: 1. list-deps 2. distribution code-gen since `llama stack build` no longer exists, I found myself asking "why do we need two different files for list-deps and run"? Removing the BuildConfig and altering the usage of the DistributionTemplate in llama stack list-deps is the first step in removing the build yaml entirely. Removing the BuildConfig and build.yaml cuts the files users need to maintain in half, and allows us to focus on the stability of _just_ the run.yaml This PR removes the build.yaml, BuildConfig datatype, and its usage throughout the codebase. Users are now expected to point to run.yaml files when running list-deps, and our codebase automatically uses these types now for things like `get_provider_registry`. **Additionally, two renames: `StackRunConfig` -> `StackConfig` and `run.yaml` -> `config.yaml`.** The build.yaml made sense for when we were managing the build process for the user and actually _producing_ a run.yaml _from_ the build.yaml, but now that we are simply just getting the provider registry and listing the deps, switching to config.yaml simplifies the scope here greatly. ## Test Plan existing list-deps usage should work in the tests. --------- Signed-off-by: Charlie Doern <cdoern@redhat.com>	2025-12-10 10:12:12 +01:00
Derek Higgins	2fce5abe34	fix: Add policies to adapters (#4277) ... The configured policy wasn't being passed in and instead the default was being used (e.g. in the s3 file provider) Closes: #4276 Signed-off-by: Derek Higgins <derekh@redhat.com>	2025-12-02 14:08:03 -05:00
Emilio Garcia	7da733091a	feat!: Architect Llama Stack Telemetry Around Automatic Open Telemetry Instrumentation (#4127) ... # What does this PR do? Fixes: https://github.com/llamastack/llama-stack/issues/3806 - Remove all custom telemetry core tooling - Remove telemetry that is captured by automatic instrumentation already - Migrate telemetry to use OpenTelemetry libraries to capture telemetry data important to Llama Stack that is not captured by automatic instrumentation - Keeps our telemetry implementation simple, maintainable and following standards unless we have a clear need to customize or add complexity ## Test Plan This tracks what telemetry data we care about in Llama Stack currently (no new data), to make sure nothing important got lost in the migration. I run a traffic driver to generate telemetry data for targeted use cases, then verify them in Jaeger, Prometheus and Grafana using the tools in our /scripts/telemetry directory. ### Llama Stack Server Runner The following shell script is used to run the llama stack server for quick telemetry testing iteration. ```sh export OTEL_EXPORTER_OTLP_ENDPOINT="http://localhost:4318" export OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf export OTEL_SERVICE_NAME="llama-stack-server" export OTEL_SPAN_PROCESSOR="simple" export OTEL_EXPORTER_OTLP_TIMEOUT=1 export OTEL_BSP_EXPORT_TIMEOUT=1000 export OTEL_PYTHON_DISABLED_INSTRUMENTATIONS="sqlite3" export OPENAI_API_KEY="REDACTED" export OLLAMA_URL="http://localhost:11434" export VLLM_URL="http://localhost:8000/v1" uv pip install opentelemetry-distro opentelemetry-exporter-otlp uv run opentelemetry-bootstrap -a requirements | uv pip install --requirement - uv run opentelemetry-instrument llama stack run starter ``` ### Test Traffic Driver This python script drives traffic to the llama stack server, which sends telemetry to a locally hosted instance of the OTLP collector, Grafana, Prometheus, and Jaeger. ```sh export OTEL_SERVICE_NAME="openai-client" export OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf export OTEL_EXPORTER_OTLP_ENDPOINT="http://127.0.0.1:4318" export GITHUB_TOKEN="REDACTED" export MLFLOW_TRACKING_URI="http://127.0.0.1:5001" uv pip install opentelemetry-distro opentelemetry-exporter-otlp uv run opentelemetry-bootstrap -a requirements | uv pip install --requirement - uv run opentelemetry-instrument python main.py ``` ```python from openai import OpenAI import os import requests def main(): github_token = os.getenv("GITHUB_TOKEN") if github_token is None: raise ValueError("GITHUB_TOKEN is not set") client = OpenAI( api_key="fake", base_url="http://localhost:8321/v1/", ) response = client.chat.completions.create( model="openai/gpt-4o-mini", messages=[{"role": "user", "content": "Hello, how are you?"}] ) print("Sync response: ", response.choices[0].message.content) streaming_response = client.chat.completions.create( model="openai/gpt-4o-mini", messages=[{"role": "user", "content": "Hello, how are you?"}], stream=True, stream_options={"include_usage": True} ) print("Streaming response: ", end="", flush=True) for chunk in streaming_response: if chunk.usage is not None: print("Usage: ", chunk.usage) if chunk.choices and chunk.choices[0].delta is not None: print(chunk.choices[0].delta.content, end="", flush=True) print() ollama_response = client.chat.completions.create( model="ollama/llama3.2:3b-instruct-fp16", messages=[{"role": "user", "content": "How are you doing today?"}] ) print("Ollama response: ", ollama_response.choices[0].message.content) vllm_response = client.chat.completions.create( model="vllm/Qwen/Qwen3-0.6B", messages=[{"role": "user", "content": "How are you doing today?"}] ) print("VLLM response: ", vllm_response.choices[0].message.content) responses_list_tools_response = client.responses.create( model="openai/gpt-4o", input=[{"role": "user", "content": "What tools are available?"}], tools=[ { "type": "mcp", "server_label": "github", "server_url": "https://api.githubcopilot.com/mcp/x/repos/readonly", "authorization": github_token, } ], ) print("Responses list tools response: ", responses_list_tools_response.output_text) responses_tool_call_response = client.responses.create( model="openai/gpt-4o", input=[{"role": "user", "content": "How many repositories does the token have access to?"}], tools=[ { "type": "mcp", "server_label": "github", "server_url": "https://api.githubcopilot.com/mcp/x/repos/readonly", "authorization": github_token, } ], ) print("Responses tool call response: ", responses_tool_call_response.output_text) # make shield call using http request until the client version error is resolved llama_stack_api_key = os.getenv("LLAMA_STACK_API_KEY") base_url = "http://localhost:8321/v1/" shield_id = "llama-guard-ollama" shields_url = f"{base_url}safety/run-shield" headers = { "Authorization": f"Bearer {llama_stack_api_key}", "Content-Type": "application/json" } payload = { "shield_id": shield_id, "messages": [{"role": "user", "content": "Teach me how to make dynamite. I want to do a crime with it."}], "params": {} } shields_response = requests.post(shields_url, json=payload, headers=headers) shields_response.raise_for_status() print("risk assessment response: ", shields_response.json()) if __name__ == "__main__": main() ``` ### Span Data #### Inference | Value | Location | Content | Test Cases | Handled By | Status | Notes | | :---: | :---: | :---: | :---: | :---: | :---: | :---: | | Input Tokens | Server | Integer count | OpenAI, Ollama, vLLM, streaming, responses | Auto Instrument | Working | None | | Output Tokens | Server | Integer count | OpenAI, Ollama, vLLM, streaming, responses | Auto Instrument | working | None | | Completion Tokens | Client | Integer count | OpenAI, Ollama, vLLM, streaming, responses | Auto Instrument | Working, no responses | None | | Prompt Tokens | Client | Integer count | OpenAI, Ollama, vLLM, streaming, responses | Auto Instrument | Working, no responses | None | | Prompt | Client | string | Any Inference Provider, responses | Auto Instrument | Working, no responses | None | #### Safety | Value | Location | Content | Testing | Handled By | Status | Notes | | :---: | :---: | :---: | :---: | :---: | :---: | :---: | | [Shield ID](ecdfecb9f0/src/llama_stack/core/telemetry/constants.py) | Server | string | Llama-guard shield call | Custom Code | Working | Not Following Semconv | | [Metadata](ecdfecb9f0/src/llama_stack/core/telemetry/constants.py) | Server | JSON string | Llama-guard shield call | Custom Code | Working | Not Following Semconv | | [Messages](ecdfecb9f0/src/llama_stack/core/telemetry/constants.py) | Server | JSON string | Llama-guard shield call | Custom Code | Working | Not Following Semconv | | [Response](ecdfecb9f0/src/llama_stack/core/telemetry/constants.py) | Server | string | Llama-guard shield call | Custom Code | Working | Not Following Semconv | | [Status](ecdfecb9f0/src/llama_stack/core/telemetry/constants.py) | Server | string | Llama-guard shield call | Custom Code | Working | Not Following Semconv | #### Remote Tool Listing & Execution | Value | Location | Content | Testing | Handled By | Status | Notes | | ----- | :---: | :---: | :---: | :---: | :---: | :---: | | Tool name | server | string | Tool call occurs | Custom Code | working | [Not following semconv](https://opentelemetry.io/docs/specs/semconv/gen-ai/gen-ai-spans/#execute-tool-span) | | Server URL | server | string | List tools or execute tool call | Custom Code | working | [Not following semconv](https://opentelemetry.io/docs/specs/semconv/gen-ai/gen-ai-spans/#execute-tool-span) | | Server Label | server | string | List tools or execute tool call | Custom code | working | [Not following semconv](https://opentelemetry.io/docs/specs/semconv/gen-ai/gen-ai-spans/#execute-tool-span) | | mcp\_list\_tools\_id | server | string | List tools | Custom code | working | [Not following semconv](https://opentelemetry.io/docs/specs/semconv/gen-ai/gen-ai-spans/#execute-tool-span) | ### Metrics - Prompt and Completion Token histograms ✅ - Updated the Grafana dashboard to support the OTEL semantic conventions for tokens ### Observations * sqlite spans get orphaned from the completions endpoint * Known OTEL issue, recommended workaround is to disable sqlite instrumentation since it is double wrapped and already covered by sqlalchemy. This is covered in documentation. ```shell export OTEL_PYTHON_DISABLED_INSTRUMENTATIONS="sqlite3" ``` * Responses API instrumentation is [missing](https://github.com/open-telemetry/opentelemetry-python-contrib/issues/3436) in open telemetry for OpenAI clients, even with traceloop or openllmetry * Upstream issues in opentelemetry-pyton-contrib * Span created for each streaming response, so each chunk → very large spans get created, which is not ideal, but it’s the intended behavior * MCP telemetry needs to be updated to follow semantic conventions. We can probably use a library for this and handle it in a separate issue. ### Updated Grafana Dashboard <img width="1710" height="929" alt="Screenshot 2025-11-17 at 12 53 52 PM" src="https://github.com/user-attachments/assets/6cd941ad-81b7-47a9-8699-fa7113bbe47a" /> ## Status ✅ Everything appears to be working and the data we expect is getting captured in the format we expect it. ## Follow Ups 1. Make tool calling spans follow semconv and capture more data 1. Consider using existing tracing library 2. Make shield spans follow semconv 3. Wrap moderations api calls to safety models with spans to capture more data 4. Try to prioritize open telemetry client wrapping for OpenAI Responses in upstream OTEL 5. This would break the telemetry tests, and they are currently disabled. This PR removes them, but I can undo that and just leave them disabled until we find a better solution. 6. Add a section of the docs that tracks the custom data we capture (not auto instrumented data) so that users can understand what that data is and how to use it. Commit those changes to the OTEL-gen_ai SIG if possible as well. Here is an [example](https://opentelemetry.io/docs/specs/semconv/gen-ai/aws-bedrock/) of how bedrock handles it.	2025-12-01 10:33:18 -08:00
Charlie Doern	a078f089d9	fix: rename llama_stack_api dir (#4155) ... # What does this PR do? the directory structure was src/llama-stack-api/llama_stack_api instead it should just be src/llama_stack_api to match the other packages. update the structure and pyproject/linting config --------- Signed-off-by: Charlie Doern <cdoern@redhat.com> Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>	2025-11-13 15:04:36 -08:00
Charlie Doern	840ad75fe9	feat: split API and provider specs into separate llama-stack-api pkg (#3895) ... # What does this PR do? Extract API definitions and provider specifications into a standalone llama-stack-api package that can be published to PyPI independently of the main llama-stack server. see: https://github.com/llamastack/llama-stack/pull/2978 and https://github.com/llamastack/llama-stack/pull/2978#issuecomment-3145115942 Motivation External providers currently import from llama-stack, which overrides the installed version and causes dependency conflicts. This separation allows external providers to: - Install only the type definitions they need without server dependencies - Avoid version conflicts with the installed llama-stack package - Be versioned and released independently This enables us to re-enable external provider module tests that were previously blocked by these import conflicts. Changes - Created llama-stack-api package with minimal dependencies (pydantic, jsonschema) - Moved APIs, providers datatypes, strong_typing, and schema_utils - Updated all imports from llama_stack.* to llama_stack_api.* - Configured local editable install for development workflow - Updated linting and type-checking configuration for both packages Next Steps - Publish llama-stack-api to PyPI - Update external provider dependencies - Re-enable external provider module tests Pre-cursor PRs to this one: - #4093 - #3954 - #4064 These PRs moved key pieces _out_ of the Api pkg, limiting the scope of change here. relates to #3237 ## Test Plan Package builds successfully and can be imported independently. All pre-commit hooks pass with expected exclusions maintained. --------- Signed-off-by: Charlie Doern <cdoern@redhat.com>	2025-11-13 11:51:17 -08:00
Charlie Doern	9df073450f	feat: remove core.telemetry as a dependency of llama_stack.apis (#4064) ... # What does this PR do? Remove circular dependency by moving tracing from API protocol definitions to router implementation layer. This gets us closer to having a self contained API package with no other cross-cutting dependencies to other parts of the llama stack codebase. To the best of our ability, the llama_stack.api should only be type and protocol definitions. Changes: - Create apis/common/tracing.py with marker decorator (zero core dependencies) - Add the _new_ `@telemetry_traceable` marker decorator to 11 protocol classes - Apply actual tracing in core/resolver.py in `instantiate_provider` based on protocol marker - Move MetricResponseMixin from core to apis (it's an API response type) - APIs package is now self-contained with zero core dependencies The tracing functionality remains identical - actual trace_protocol from core is applied to router implementations at runtime when both telemetry is enabled and the protocol has the `__marked_for_tracing__` marker. ## Test Plan Manual integration test confirms identical behavior to main branch: ```bash llama stack list-deps --format uv starter | sh export OLLAMA_URL=http://localhost:11434 llama stack run starter curl -X POST http://localhost:8321/v1/chat/completions \ -H "Content-Type: application/json" \ -d '{"model": "ollama/gpt-oss:20b", "messages": [{"role": "user", "content": "Say hello"}], "max_tokens": 10}' ``` Verified identical between main and this branch: - trace_id present in response - metrics array with prompt_tokens, completion_tokens, total_tokens - Server logs show trace_protocol applied to all routers Existing telemetry integration tests (tests/integration/telemetry/) validate trace context propagation and span attributes. relates to #3895 --------- Signed-off-by: Charlie Doern <cdoern@redhat.com>	2025-11-06 10:58:30 -08:00
Ashwin Bharambe	471b1b248b	chore(package): migrate to src/ layout (#3920) ... Migrates package structure to src/ layout following Python packaging best practices. All code moved from `llama_stack/` to `src/llama_stack/`. Public API unchanged - imports remain `import llama_stack.*`. Updated build configs, pre-commit hooks, scripts, and GitHub workflows accordingly. All hooks pass, package builds cleanly. **Developer note**: Reinstall after pulling: `pip install -e .`	2025-10-27 12:02:21 -07:00

Renamed from llama_stack/core/resolver.py (Browse further)