Merge branch 'main' into pr2088

.github/workflows/providers-build.yml

@@ -153,7 +153,7 @@ jobs:
         uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
 
       - name: Set up Python
-        uses: actions/setup-python@8d9ed9ac5c53483de85588cdf95a591a75ab9f55 # v5.5.0
+        uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # v5.6.0
         with:
           python-version: '3.10'
 

CONTRIBUTING.md

@@ -141,11 +141,18 @@ uv sync
 
 ## Coding Style
 
-* Comments should provide meaningful insights into the code. Avoid filler comments that simply describe the next step, as they create unnecessary clutter, same goes for docstrings.
+* Comments should provide meaningful insights into the code. Avoid filler comments that simply
-* Prefer comments to clarify surprising behavior and/or relationships between parts of the code rather than explain what the next line of code does.
+  describe the next step, as they create unnecessary clutter, same goes for docstrings.
-* Catching exceptions, prefer using a specific exception type rather than a broad catch-all like `Exception`.
+* Prefer comments to clarify surprising behavior and/or relationships between parts of the code
+  rather than explain what the next line of code does.
+* Catching exceptions, prefer using a specific exception type rather than a broad catch-all like
+  `Exception`.
 * Error messages should be prefixed with "Failed to ..."
-* 4 spaces for indentation rather than tabs
+* 4 spaces for indentation rather than tab
+* When using `# noqa` to suppress a style or linter warning, include a comment explaining the
+  justification for bypassing the check.
+* When using `# type: ignore` to suppress a mypy warning, include a comment explaining the
+  justification for bypassing the check.
 
 ## Common Tasks
 

docs/_static/css/my_theme.css

@@ -27,3 +27,9 @@ pre {
     white-space: pre-wrap !important;
     word-break: break-all;
 }
+
+[data-theme="dark"] .mermaid {
+    background-color: #f4f4f6 !important;
+    border-radius: 6px;
+    padding: 0.5em;
+  }

docs/source/providers/external.md

@@ -53,6 +53,7 @@ Here's a list of known external providers that you can use with Llama Stack:
 | Name | Description | API | Type | Repository |
 |------|-------------|-----|------|------------|
 | KubeFlow Training | Train models with KubeFlow | Post Training | Remote | [llama-stack-provider-kft](https://github.com/opendatahub-io/llama-stack-provider-kft) |
+| KubeFlow Pipelines | Train models with KubeFlow Pipelines | Post Training | Remote | [llama-stack-provider-kfp-trainer](https://github.com/opendatahub-io/llama-stack-provider-kfp-trainer) |
 | RamaLama | Inference models with RamaLama | Inference | Remote | [ramalama-stack](https://github.com/containers/ramalama-stack) |
 
 ### Remote Provider Specification

install.sh

@@ -16,61 +16,120 @@ WAIT_TIMEOUT=300
 log(){ printf "\e[1;32m%s\e[0m\n" "$*"; }
 die(){ printf "\e[1;31m❌ %s\e[0m\n" "$*" >&2; exit 1; }
 
+wait_for_service() {
+  local url="$1"
+  local pattern="$2"
+  local timeout="$3"
+  local name="$4"
+  local start ts
+  log "⏳  Waiting for ${name}…"
+  start=$(date +%s)
+  while true; do
+    if curl --retry 5 --retry-delay 1 --retry-max-time "$timeout" --retry-all-errors --silent --fail "$url" 2>/dev/null | grep -q "$pattern"; then
+      break
+    fi
+    ts=$(date +%s)
+    if (( ts - start >= timeout )); then
+      return 1
+    fi
+    printf '.'
+    sleep 1
+  done
+  return 0
+}
+
 if command -v docker &> /dev/null; then
   ENGINE="docker"
-  HOST_DNS="host.docker.internal"
 elif command -v podman &> /dev/null; then
   ENGINE="podman"
-  HOST_DNS="host.containers.internal"
 else
   die "Docker or Podman is required. Install Docker: https://docs.docker.com/get-docker/ or Podman: https://podman.io/getting-started/installation"
 fi
 
+# Explicitly set the platform for the host architecture
+HOST_ARCH="$(uname -m)"
+if [ "$HOST_ARCH" = "arm64" ]; then
+  if [ "$ENGINE" = "docker" ]; then
+    PLATFORM_OPTS=( --platform linux/amd64 )
+  else
+    PLATFORM_OPTS=( --os linux --arch amd64 )
+  fi
+else
+  PLATFORM_OPTS=()
+fi
+
+# macOS + Podman: ensure VM is running before we try to launch containers
+# If you need GPU passthrough under Podman on macOS, init the VM with libkrun:
+#   CONTAINERS_MACHINE_PROVIDER=libkrun podman machine init
+if [ "$ENGINE" = "podman" ] && [ "$(uname -s)" = "Darwin" ]; then
+  if ! podman info &>/dev/null; then
+    log "⌛️ Initializing Podman VM…"
+    podman machine init &>/dev/null || true
+    podman machine start &>/dev/null || true
+
+    log "⌛️  Waiting for Podman API…"
+    until podman info &>/dev/null; do
+      sleep 1
+    done
+    log "✅  Podman VM is up"
+  fi
+fi
+
 # Clean up any leftovers from earlier runs
 for name in ollama-server llama-stack; do
   ids=$($ENGINE ps -aq --filter "name=^${name}$")
   if [ -n "$ids" ]; then
-    log "⚠️   Found existing container(s) for '${name}', removing..."
+    log "⚠️   Found existing container(s) for '${name}', removing…"
-    $ENGINE rm -f "$ids"
+    $ENGINE rm -f "$ids" > /dev/null 2>&1
   fi
 done
 
+###############################################################################
+# 0. Create a shared network
+###############################################################################
+if ! $ENGINE network inspect llama-net >/dev/null 2>&1; then
+  log "🌐  Creating network…"
+  $ENGINE network create llama-net >/dev/null 2>&1
+fi
+
 ###############################################################################
 # 1. Ollama
 ###############################################################################
 log "🦙  Starting Ollama…"
-$ENGINE run -d --name ollama-server \
+$ENGINE run -d "${PLATFORM_OPTS[@]}" --name ollama-server \
-  -p "${OLLAMA_PORT}:11434" \
+  --network llama-net \
+  -p "${OLLAMA_PORT}:${OLLAMA_PORT}" \
   ollama/ollama > /dev/null 2>&1
 
-log "⏳  Waiting for Ollama daemon…"
+if ! wait_for_service "http://localhost:${OLLAMA_PORT}/" "Ollama" "$WAIT_TIMEOUT" "Ollama daemon"; then
-if ! timeout "$WAIT_TIMEOUT" bash -c \
-    "until curl -fsS http://localhost:${OLLAMA_PORT}/ 2>/dev/null | grep -q 'Ollama'; do sleep 1; done"; then
   log "❌  Ollama daemon did not become ready in ${WAIT_TIMEOUT}s; dumping container logs:"
-  $ENGINE logs ollama-server --tail=200
+  $ENGINE logs --tail 200 ollama-server
   die "Ollama startup failed"
 fi
 
-log "📦  Ensuring model is pulled: ${MODEL_ALIAS}..."
+log "📦  Ensuring model is pulled: ${MODEL_ALIAS}…"
-$ENGINE exec ollama-server ollama pull "${MODEL_ALIAS}" > /dev/null 2>&1
+if ! $ENGINE exec ollama-server ollama pull "${MODEL_ALIAS}" > /dev/null 2>&1; then
+  log "❌  Failed to pull model ${MODEL_ALIAS}; dumping container logs:"
+  $ENGINE logs --tail 200 ollama-server
+  die "Model pull failed"
+fi
 
 ###############################################################################
 # 2. Llama‑Stack
 ###############################################################################
-log "🦙📦  Starting Llama‑Stack…"
+cmd=( run -d "${PLATFORM_OPTS[@]}" --name llama-stack \
-$ENGINE run -d --name llama-stack \
+      --network llama-net \
       -p "${PORT}:${PORT}" \
-  --add-host="${HOST_DNS}:host-gateway" \
+      "${SERVER_IMAGE}" --port "${PORT}" \
-  "${SERVER_IMAGE}" \
-  --port "${PORT}" \
       --env INFERENCE_MODEL="${MODEL_ALIAS}" \
-  --env OLLAMA_URL="http://${HOST_DNS}:${OLLAMA_PORT}" > /dev/null 2>&1
+      --env OLLAMA_URL="http://ollama-server:${OLLAMA_PORT}" )
 
-log "⏳  Waiting for Llama-Stack API…"
+log "🦙  Starting Llama‑Stack…"
-if ! timeout "$WAIT_TIMEOUT" bash -c \
+$ENGINE "${cmd[@]}" > /dev/null 2>&1
-  "until curl -fsS http://localhost:${PORT}/v1/health 2>/dev/null | grep -q 'OK'; do sleep 1; done"; then
+
+if ! wait_for_service "http://127.0.0.1:${PORT}/v1/health" "OK" "$WAIT_TIMEOUT" "Llama-Stack API"; then
   log "❌  Llama-Stack did not become ready in ${WAIT_TIMEOUT}s; dumping container logs:"
-  $ENGINE logs llama-stack --tail=200
+  $ENGINE logs --tail 200 llama-stack
   die "Llama-Stack startup failed"
 fi
 

llama_stack/providers/utils/inference/openai_compat.py

@@ -108,6 +108,7 @@ from llama_stack.apis.inference.inference import (
     OpenAIChatCompletion,
     OpenAICompletion,
     OpenAICompletionChoice,
+    OpenAIMessageParam,
     OpenAIResponseFormatParam,
     ToolConfig,
 )
@@ -987,7 +988,7 @@ def _convert_openai_sampling_params(
 
 
 def openai_messages_to_messages(
-    messages: list[OpenAIChatCompletionMessage],
+    messages: list[OpenAIMessageParam],
 ) -> list[Message]:
     """
     Convert a list of OpenAIChatCompletionMessage into a list of Message.
@@ -995,12 +996,12 @@ def openai_messages_to_messages(
     converted_messages = []
     for message in messages:
         if message.role == "system":
-            converted_message = SystemMessage(content=message.content)
+            converted_message = SystemMessage(content=openai_content_to_content(message.content))
         elif message.role == "user":
             converted_message = UserMessage(content=openai_content_to_content(message.content))
         elif message.role == "assistant":
             converted_message = CompletionMessage(
-                content=message.content,
+                content=openai_content_to_content(message.content),
                 tool_calls=_convert_openai_tool_calls(message.tool_calls),
                 stop_reason=StopReason.end_of_turn,
             )
@@ -1331,7 +1332,7 @@ class OpenAIChatCompletionToLlamaStackMixin:
     async def openai_chat_completion(
         self,
         model: str,
-        messages: list[OpenAIChatCompletionMessage],
+        messages: list[OpenAIMessageParam],
         frequency_penalty: float | None = None,
         function_call: str | dict[str, Any] | None = None,
         functions: list[dict[str, Any]] | None = None,

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "llama_stack"
-version = "0.2.4"
+version = "0.2.5"
 authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
 description = "Llama Stack"
 readme = "README.md"
@@ -27,7 +27,7 @@ dependencies = [
     "huggingface-hub",
     "jinja2>=3.1.6",
     "jsonschema",
-    "llama-stack-client>=0.2.4",
+    "llama-stack-client>=0.2.5",
     "openai>=1.66",
     "prompt-toolkit",
     "python-dotenv",
@@ -105,7 +105,7 @@ codegen = ["rich", "pydantic", "jinja2>=3.1.6"]
 ui = [
     "streamlit",
     "pandas",
-    "llama-stack-client>=0.2.4",
+    "llama-stack-client>=0.2.5",
     "streamlit-option-menu",
 ]
 

requirements.txt

@@ -26,7 +26,7 @@ jiter==0.8.2
 jsonschema==4.23.0
 jsonschema-specifications==2024.10.1
 kubernetes==32.0.1
-llama-stack-client==0.2.4
+llama-stack-client==0.2.5
 lxml==5.3.1
 markdown-it-py==3.0.0
 markupsafe==3.0.2

tests/integration/agents/test_agents.py

@@ -548,6 +548,7 @@ def test_rag_agent_with_attachments(llama_stack_client, agent_config):
     assert "lora" in response.output_message.content.lower()
 
 
+@pytest.mark.skip(reason="Code interpreter is currently disabled in the Stack")
 def test_rag_and_code_agent(llama_stack_client, agent_config):
     if "llama-4" in agent_config["model"].lower():
         pytest.xfail("Not working for llama4")

tests/unit/providers/utils/inference/test_openai_compat.py

@@ -7,9 +7,20 @@
 import pytest
 
 from llama_stack.apis.common.content_types import TextContentItem
-from llama_stack.apis.inference.inference import CompletionMessage, UserMessage
+from llama_stack.apis.inference.inference import (
+    CompletionMessage,
+    OpenAIAssistantMessageParam,
+    OpenAIChatCompletionContentPartTextParam,
+    OpenAISystemMessageParam,
+    OpenAIUserMessageParam,
+    SystemMessage,
+    UserMessage,
+)
 from llama_stack.models.llama.datatypes import BuiltinTool, StopReason, ToolCall
-from llama_stack.providers.utils.inference.openai_compat import convert_message_to_openai_dict
+from llama_stack.providers.utils.inference.openai_compat import (
+    convert_message_to_openai_dict,
+    openai_messages_to_messages,
+)
 
 
 @pytest.mark.asyncio
@@ -67,3 +78,39 @@ async def test_convert_message_to_openai_dict_with_builtin_tool_call():
             {"id": "123", "type": "function", "function": {"name": "brave_search", "arguments": '{"foo": "bar"}'}}
         ],
     }
+
+
+@pytest.mark.asyncio
+async def test_openai_messages_to_messages_with_content_str():
+    openai_messages = [
+        OpenAISystemMessageParam(content="system message"),
+        OpenAIUserMessageParam(content="user message"),
+        OpenAIAssistantMessageParam(content="assistant message"),
+    ]
+
+    llama_messages = openai_messages_to_messages(openai_messages)
+    assert len(llama_messages) == 3
+    assert isinstance(llama_messages[0], SystemMessage)
+    assert isinstance(llama_messages[1], UserMessage)
+    assert isinstance(llama_messages[2], CompletionMessage)
+    assert llama_messages[0].content == "system message"
+    assert llama_messages[1].content == "user message"
+    assert llama_messages[2].content == "assistant message"
+
+
+@pytest.mark.asyncio
+async def test_openai_messages_to_messages_with_content_list():
+    openai_messages = [
+        OpenAISystemMessageParam(content=[OpenAIChatCompletionContentPartTextParam(text="system message")]),
+        OpenAIUserMessageParam(content=[OpenAIChatCompletionContentPartTextParam(text="user message")]),
+        OpenAIAssistantMessageParam(content=[OpenAIChatCompletionContentPartTextParam(text="assistant message")]),
+    ]
+
+    llama_messages = openai_messages_to_messages(openai_messages)
+    assert len(llama_messages) == 3
+    assert isinstance(llama_messages[0], SystemMessage)
+    assert isinstance(llama_messages[1], UserMessage)
+    assert isinstance(llama_messages[2], CompletionMessage)
+    assert llama_messages[0].content[0].text == "system message"
+    assert llama_messages[1].content[0].text == "user message"
+    assert llama_messages[2].content[0].text == "assistant message"