Files
TradingAgents/tradingagents/llm_clients/capabilities.py
T
Yijia-Xiao 22bb91bd83 fix(llm): structured output for DeepSeek V4 and reasoner
DeepSeek V4 and reasoner reject tool_choice but accept tools.
Route via a per-model capability table that suppresses tool_choice
for thinking-mode models.

#678 #689
2026-05-11 01:12:28 +00:00

96 lines
3.4 KiB
Python

"""Declarative per-model capability table for OpenAI-compatible providers.
This is the single place that knows which model IDs reject which API
parameters or require which structured-output method. The LLM client
subclasses consult ``get_capabilities(model_name)`` instead of hardcoding
model-name ``if`` ladders, so adding a new model (or a new provider quirk)
means editing this table — not the client code.
Pattern adapted from the per-model ``compat:`` flags DeepSeek themselves
publish in their integration guides (e.g. the Oh My Pi config schema
documents ``supportsToolChoice``, ``requiresReasoningContentForToolCalls``
as declarative per-model fields).
"""
from __future__ import annotations
import re
from dataclasses import dataclass
from typing import Literal
StructuredMethod = Literal[
"function_calling", # uses tools; respects supports_tool_choice
"json_mode", # uses response_format={"type":"json_object"}
"json_schema", # uses response_format={"type":"json_schema",...}
"none", # no structured output available; caller falls back to free-text
]
@dataclass(frozen=True)
class ModelCapabilities:
"""What an OpenAI-compatible model accepts at the API level."""
supports_tool_choice: bool
supports_json_mode: bool
supports_json_schema: bool
preferred_structured_method: StructuredMethod
# DeepSeek thinking-mode models 400 if reasoning_content from prior
# assistant turns is not echoed back on the next request.
requires_reasoning_content_roundtrip: bool = False
# DeepSeek's thinking models accept the ``tools`` array but reject the
# ``tool_choice`` parameter (official Oh My Pi integration guide and the
# 400 response in issue #678). Their official tool-calling examples
# (api-docs.deepseek.com/guides/tool_calls) pass ``tools=[...]`` without
# ``tool_choice`` — we mirror that pattern by setting supports_tool_choice
# to False and letting the client suppress the kwarg.
_DEEPSEEK_THINKING = ModelCapabilities(
supports_tool_choice=False,
supports_json_mode=True,
supports_json_schema=False,
preferred_structured_method="function_calling",
requires_reasoning_content_roundtrip=True,
)
_DEEPSEEK_CHAT = ModelCapabilities(
supports_tool_choice=True,
supports_json_mode=True,
supports_json_schema=False,
preferred_structured_method="function_calling",
)
_DEFAULT = ModelCapabilities(
supports_tool_choice=True,
supports_json_mode=True,
supports_json_schema=True,
preferred_structured_method="function_calling",
)
# Exact-ID matches take precedence over pattern matches.
_BY_ID: dict[str, ModelCapabilities] = {
"deepseek-chat": _DEEPSEEK_CHAT,
"deepseek-reasoner": _DEEPSEEK_THINKING,
"deepseek-v4-flash": _DEEPSEEK_THINKING,
"deepseek-v4-pro": _DEEPSEEK_THINKING,
}
# Forward-compat patterns. A new ``deepseek-v5-*`` or ``deepseek-reasoner-*``
# variant inherits the thinking-mode quirks automatically.
_BY_PATTERN: list[tuple[re.Pattern[str], ModelCapabilities]] = [
(re.compile(r"^deepseek-v\d"), _DEEPSEEK_THINKING),
(re.compile(r"^deepseek-reasoner"), _DEEPSEEK_THINKING),
]
def get_capabilities(model_name: str) -> ModelCapabilities:
"""Resolve capabilities by exact ID, then pattern, then default."""
if model_name in _BY_ID:
return _BY_ID[model_name]
for pattern, caps in _BY_PATTERN:
if pattern.match(model_name):
return caps
return _DEFAULT