vllm-cpu 0.11.0.post2__cp312-cp312-manylinux_2_17_x86_64.whl

vllm/tracing.py

@@ -0,0 +1,136 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+
+import os
+from collections.abc import Mapping
+from typing import Optional
+
+from vllm.logger import init_logger
+from vllm.utils import run_once
+
+TRACE_HEADERS = ["traceparent", "tracestate"]
+
+logger = init_logger(__name__)
+
+_is_otel_imported = False
+otel_import_error_traceback: Optional[str] = None
+try:
+    from opentelemetry.context.context import Context
+    from opentelemetry.sdk.environment_variables import (
+        OTEL_EXPORTER_OTLP_TRACES_PROTOCOL)
+    from opentelemetry.sdk.trace import TracerProvider
+    from opentelemetry.sdk.trace.export import BatchSpanProcessor
+    from opentelemetry.trace import SpanKind, Tracer, set_tracer_provider
+    from opentelemetry.trace.propagation.tracecontext import (
+        TraceContextTextMapPropagator)
+    _is_otel_imported = True
+except ImportError:
+    # Capture and format traceback to provide detailed context for the import
+    # error. Only the string representation of the error is retained to avoid
+    # memory leaks.
+    # See https://github.com/vllm-project/vllm/pull/7266#discussion_r1707395458
+    import traceback
+    otel_import_error_traceback = traceback.format_exc()
+
+    class Context:  # type: ignore
+        pass
+
+    class BaseSpanAttributes:  # type: ignore
+        pass
+
+    class SpanKind:  # type: ignore
+        pass
+
+    class Tracer:  # type: ignore
+        pass
+
+
+def is_otel_available() -> bool:
+    return _is_otel_imported
+
+
+def init_tracer(instrumenting_module_name: str,
+                otlp_traces_endpoint: str) -> Optional[Tracer]:
+    if not is_otel_available():
+        raise ValueError(
+            "OpenTelemetry is not available. Unable to initialize "
+            "a tracer. Ensure OpenTelemetry packages are installed. "
+            f"Original error:\n{otel_import_error_traceback}")
+    trace_provider = TracerProvider()
+
+    span_exporter = get_span_exporter(otlp_traces_endpoint)
+    trace_provider.add_span_processor(BatchSpanProcessor(span_exporter))
+    set_tracer_provider(trace_provider)
+
+    tracer = trace_provider.get_tracer(instrumenting_module_name)
+    return tracer
+
+
+def get_span_exporter(endpoint):
+    protocol = os.environ.get(OTEL_EXPORTER_OTLP_TRACES_PROTOCOL, "grpc")
+    if protocol == "grpc":
+        from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import (
+            OTLPSpanExporter)
+    elif protocol == "http/protobuf":
+        from opentelemetry.exporter.otlp.proto.http.trace_exporter import (
+            OTLPSpanExporter)  # type: ignore
+    else:
+        raise ValueError(
+            f"Unsupported OTLP protocol '{protocol}' is configured")
+
+    return OTLPSpanExporter(endpoint=endpoint)
+
+
+def extract_trace_context(
+        headers: Optional[Mapping[str, str]]) -> Optional[Context]:
+    if is_otel_available():
+        headers = headers or {}
+        return TraceContextTextMapPropagator().extract(headers)
+    else:
+        return None
+
+
+def extract_trace_headers(headers: Mapping[str, str]) -> Mapping[str, str]:
+
+    return {h: headers[h] for h in TRACE_HEADERS if h in headers}
+
+
+class SpanAttributes:
+    # Attribute names copied from here to avoid version conflicts:
+    # https://github.com/open-telemetry/semantic-conventions/blob/main/docs/gen-ai/gen-ai-spans.md
+    GEN_AI_USAGE_COMPLETION_TOKENS = "gen_ai.usage.completion_tokens"
+    GEN_AI_USAGE_PROMPT_TOKENS = "gen_ai.usage.prompt_tokens"
+    GEN_AI_REQUEST_MAX_TOKENS = "gen_ai.request.max_tokens"
+    GEN_AI_REQUEST_TOP_P = "gen_ai.request.top_p"
+    GEN_AI_REQUEST_TEMPERATURE = "gen_ai.request.temperature"
+    GEN_AI_RESPONSE_MODEL = "gen_ai.response.model"
+    # Attribute names added until they are added to the semantic conventions:
+    GEN_AI_REQUEST_ID = "gen_ai.request.id"
+    GEN_AI_REQUEST_N = "gen_ai.request.n"
+    GEN_AI_USAGE_NUM_SEQUENCES = "gen_ai.usage.num_sequences"
+    GEN_AI_LATENCY_TIME_IN_QUEUE = "gen_ai.latency.time_in_queue"
+    GEN_AI_LATENCY_TIME_TO_FIRST_TOKEN = "gen_ai.latency.time_to_first_token"
+    GEN_AI_LATENCY_E2E = "gen_ai.latency.e2e"
+    GEN_AI_LATENCY_TIME_IN_SCHEDULER = "gen_ai.latency.time_in_scheduler"
+    # Time taken in the forward pass for this across all workers
+    GEN_AI_LATENCY_TIME_IN_MODEL_FORWARD = (
+        "gen_ai.latency.time_in_model_forward")
+    # Time taken in the model execute function. This will include model
+    # forward, block/sync across workers, cpu-gpu sync time and sampling time.
+    GEN_AI_LATENCY_TIME_IN_MODEL_EXECUTE = (
+        "gen_ai.latency.time_in_model_execute")
+    GEN_AI_LATENCY_TIME_IN_MODEL_PREFILL = \
+        "gen_ai.latency.time_in_model_prefill"
+    GEN_AI_LATENCY_TIME_IN_MODEL_DECODE = "gen_ai.latency.time_in_model_decode"
+    GEN_AI_LATENCY_TIME_IN_MODEL_INFERENCE = \
+        "gen_ai.latency.time_in_model_inference"
+
+
+def contains_trace_headers(headers: Mapping[str, str]) -> bool:
+    return any(h in headers for h in TRACE_HEADERS)
+
+
+@run_once
+def log_tracing_disabled_warning() -> None:
+    logger.warning(
+        "Received a request with trace context but tracing is disabled")

vllm/transformers_utils/__init__.py

@@ -0,0 +1,24 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+
+from vllm import envs
+
+if envs.VLLM_USE_MODELSCOPE:
+    try:
+        # Patch here, before each import happens
+        import modelscope
+        from packaging import version
+
+        # patch_hub begins from modelscope>=1.18.1
+        if version.parse(modelscope.__version__) <= version.parse('1.18.0'):
+            raise ImportError(
+                'Using vLLM with ModelScope needs modelscope>=1.18.1, please '
+                'install by `pip install modelscope -U`')
+        from modelscope.utils.hf_util import patch_hub
+
+        # Patch hub to download models from modelscope to speed up.
+        patch_hub()
+    except ImportError as err:
+        raise ImportError(
+            "Please install modelscope>=1.18.1 via "
+            "`pip install modelscope>=1.18.1` to use ModelScope.") from err

vllm/transformers_utils/chat_templates/__init__.py

@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+from .registry import get_chat_template_fallback_path
+
+__all__ = ["get_chat_template_fallback_path"]

vllm/transformers_utils/chat_templates/registry.py

@@ -0,0 +1,70 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+from pathlib import Path
+from typing import Callable, Optional, Union
+
+from vllm.logger import init_logger
+
+logger = init_logger(__file__)
+
+CHAT_TEMPLATES_DIR = Path(__file__).parent
+
+ChatTemplatePath = Union[Path, Callable[[str], Optional[Path]]]
+
+
+def _get_qwen_chat_template_fallback(
+        tokenizer_name_or_path: str) -> Optional[Path]:
+    if tokenizer_name_or_path.endswith("-Chat"):
+        return CHAT_TEMPLATES_DIR / "template_chatml.jinja"
+
+    return CHAT_TEMPLATES_DIR / "template_basic.jinja"
+
+
+def _get_minicpmv_chat_template_fallback(
+        tokenizer_name_or_path: str) -> Optional[Path]:
+    # MiniCPM-V-4.5 version uses a dedicated template
+    if "4.5" in tokenizer_name_or_path or "4_5" in tokenizer_name_or_path:
+        return CHAT_TEMPLATES_DIR / "template_minicpmv45.jinja"
+
+    # Other versions use chatml template
+    return CHAT_TEMPLATES_DIR / "template_chatml.jinja"
+
+
+# yapf: disable
+_MODEL_TYPE_TO_CHAT_TEMPLATE_FALLBACK: dict[str, ChatTemplatePath] = {
+    "blip-2": CHAT_TEMPLATES_DIR / "template_blip2.jinja",
+    "chameleon": CHAT_TEMPLATES_DIR / "template_basic.jinja",
+    "deepseek_vl_v2": CHAT_TEMPLATES_DIR / "template_deepseek_vl2.jinja",
+    "fuyu": CHAT_TEMPLATES_DIR / "template_fuyu.jinja",
+    "minicpmv": _get_minicpmv_chat_template_fallback,
+    "paligemma": CHAT_TEMPLATES_DIR / "template_basic.jinja",
+    "qwen": _get_qwen_chat_template_fallback,
+}
+# yapf: enable
+
+
+def register_chat_template_fallback_path(
+    model_type: str,
+    chat_template: ChatTemplatePath,
+) -> None:
+    if model_type in _MODEL_TYPE_TO_CHAT_TEMPLATE_FALLBACK:
+        logger.warning(
+            "Model type %s already has a chat template registered. "
+            "It will be overwritten by the new chat template %s.", model_type,
+            chat_template)
+
+    _MODEL_TYPE_TO_CHAT_TEMPLATE_FALLBACK[model_type] = chat_template
+
+
+def get_chat_template_fallback_path(
+    model_type: str,
+    tokenizer_name_or_path: str,
+) -> Optional[Path]:
+    chat_template = _MODEL_TYPE_TO_CHAT_TEMPLATE_FALLBACK.get(model_type)
+    if callable(chat_template):
+        chat_template = chat_template(tokenizer_name_or_path)
+
+    if chat_template is None:
+        return None
+
+    return chat_template

vllm/transformers_utils/chat_templates/template_basic.jinja

@@ -0,0 +1,3 @@
+{%- for message in messages -%}
+    {{- message['content'] -}}
+{%- endfor -%}

vllm/transformers_utils/chat_templates/template_blip2.jinja

@@ -0,0 +1,11 @@
+{%- for message in messages -%}
+    {%- if message['role'] == 'user' -%}
+        {{- 'Question: ' + message['content'] + ' ' -}}
+    {%- elif message['role'] == 'assistant' -%}
+        {{- 'Answer: ' + message['content'] + ' ' -}}
+    {%- endif -%}
+{%- endfor -%}
+
+{%- if add_generation_prompt -%}
+    {{- 'Answer:' -}}
+{% endif %}

vllm/transformers_utils/chat_templates/template_chatml.jinja

@@ -0,0 +1,10 @@
+{%- for message in messages -%}
+    {{- '<|im_start|>' + message['role'] + '\n' + message['content'] -}}
+    {%- if (loop.last and add_generation_prompt) or not loop.last -%}
+        {{- '<|im_end|>' + '\n' -}}
+    {%- endif -%}
+{%- endfor -%}
+
+{%- if add_generation_prompt and messages[-1]['role'] != 'assistant' -%}
+    {{- '<|im_start|>assistant\n' -}}
+{%- endif -%}

vllm/transformers_utils/chat_templates/template_deepseek_vl2.jinja

@@ -0,0 +1,23 @@
+{%- if messages[0]['role'] == 'system' -%}
+    {%- set system_message = messages[0]['content'] -%}
+    {%- set messages = messages[1:] -%}
+{%- else -%}
+    {% set system_message = '' -%}
+{%- endif -%}
+
+{{ bos_token + system_message }}
+{%- for message in messages -%}
+    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
+        {{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}
+    {%- endif -%}
+
+    {%- if message['role'] == 'user' -%}
+        {{ '<|User|>: ' + message['content'] + '\n\n' }}
+    {%- elif message['role'] == 'assistant' -%}
+        {{ '<|Assistant|>: ' + message['content'] + eos_token + '\n\n' }}
+    {%- endif -%}
+{%- endfor -%}
+
+{%- if add_generation_prompt -%}
+    {{ '<|Assistant|>: ' }}
+{%- endif -%}

vllm/transformers_utils/chat_templates/template_fuyu.jinja

@@ -0,0 +1,3 @@
+{%- for message in messages -%}
+    {{- message['content'] + '\n' -}}
+{%- endfor -%}

vllm/transformers_utils/chat_templates/template_minicpmv45.jinja

@@ -0,0 +1,93 @@
+{%- set enable_thinking = enable_thinking | default(false) %}
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+    {%- if enable_thinking is defined and enable_thinking is true %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}