olgagaga 0ca0fe2221 fix(providers): wire MiMo thinking control on gateway providers (#3845) ...

The xiaomi_mimo ProviderSpec carries thinking_style="thinking_type", but
gateway providers (OpenRouter etc.) route MiMo under their own spec
which has no thinking_style. As a result, `reasoning_effort="none"` was
silently ignored: `{"thinking": {"type": "disabled"}}` was never
injected and responses still contained reasoning_content.

Mirror the Kimi pattern that already handles the same problem: add an
explicit _MIMO_THINKING_MODELS allowlist (mimo-v2.5-pro, mimo-v2.5,
mimo-v2-pro, mimo-v2-omni — per Xiaomi docs), an _is_mimo_thinking_model
helper that strips publisher prefixes ("xiaomi/mimo-v2.5-pro" matches),
and a sibling branch in _build_kwargs that injects the thinking payload
by model name. mimo-v2-flash is intentionally excluded — it has no
thinking mode.

Also include MiMo in the explicit_thinking predicate so the
reasoning_content backfill (#3554, #3584) covers the gateway path
consistently with the direct path.

Tests cover the gateway disable/enable signals, bare-slug fallback,
flash exclusion, and a non-MiMo sanity check.

2026-05-16 20:46:34 +08:00

..

test_anthropic_long_request_fallback.py

test(provider): tidy Anthropic fallback imports

2026-05-01 23:59:24 +08:00

test_anthropic_merge_consecutive.py

fix(anthropic): also enforce leading-user + empty-array recovery

2026-04-21 01:32:32 +08:00

test_anthropic_stream_idle.py

feat(goal): /goal command & long-running tasks (long_task)

2026-05-16 01:14:11 +08:00

test_anthropic_thinking.py

test(providers): cover reasoning_effort="none" and gemma auto-routing

2026-04-29 15:41:11 +08:00

test_anthropic_tool_result.py

test(anthropic): pin tool_result image_url conversion regression

2026-04-22 22:10:53 +08:00

test_azure_openai_provider.py

test(providers): cover reasoning_effort="none" and gemma auto-routing

2026-04-29 15:41:11 +08:00

test_bedrock_provider.py

fix(provider): preserve Bedrock tool config for history

2026-05-12 20:59:01 +08:00

test_cached_tokens.py

fix(providers): normalize anthropic cached token usage

2026-04-02 12:51:45 +08:00

test_custom_provider.py

fix(provider): preserve static error helper compatibility

2026-04-13 09:37:31 +08:00

test_enforce_role_alternation.py

refactor(agent): tighten comments, extract constant, strengthen edge case test

2026-04-17 16:20:53 +08:00

test_extra_body_config.py

feat(providers): add extra_body config for OpenAI-compatible endpoints

2026-04-28 15:56:13 +08:00

test_github_copilot_routing.py

test(providers): add regression tests for GitHub Copilot /responses routing

2026-04-22 06:53:37 +00:00

test_image_generation.py

feat: add image generation tool and WebUI mode

2026-05-08 20:06:23 +08:00

test_litellm_kwargs.py

feat(goal): /goal command & long-running tasks (long_task)

2026-05-16 01:14:11 +08:00

test_llm_response.py

test + docs: lock should_execute_tools guard semantics (#3220)

2026-04-17 20:39:46 +08:00

test_local_endpoint_detection.py

fix(providers): tighten local endpoint detection

2026-04-26 16:14:24 +08:00

test_longcat_provider.py

fix(provider): correct LongCat OpenAI base URL

2026-05-02 01:52:04 +08:00

test_minimax_anthropic_provider.py

docs(providers): explain MiniMax thinking endpoint

2026-04-16 01:00:45 +08:00

test_mistral_provider.py

refactor: replace litellm with native openai + anthropic SDKs

2026-03-25 01:58:48 +08:00

test_openai_codex_provider.py

fix(codex): stabilize prompt cache key

2026-05-16 00:13:10 +08:00

test_openai_compat_timeout.py

fix(provider): bound OpenAI-compatible request timeouts

2026-04-27 17:47:31 +08:00

test_openai_responses.py

refactor(providers): rename openai responses helpers

2026-04-02 13:43:34 +08:00

test_prompt_cache_markers.py

fix(cache): stabilize tool ordering and cache markers for MCP

2026-04-01 17:07:22 +00:00

test_provider_error_metadata.py

refactor: remove dead _error_response wrapper in Anthropic provider

2026-04-06 08:35:02 +00:00

test_provider_retry_after_hints.py

fix: robust Retry-After extraction across provider backends

2026-04-02 18:39:24 +00:00

test_provider_retry.py

fix(retry): recognize ZhiPu 1302 rate-limit error for retry

2026-04-21 21:23:20 +08:00

test_provider_sdk_retry_defaults.py

fix(providers): disable Azure SDK retries by default

2026-04-04 12:36:45 +08:00

test_providers_init.py

add native Bedrock Converse provider

2026-05-01 18:52:03 +08:00

test_reasoning_content.py

test(providers): cover reasoning_content extraction in OpenAI compat provider

2026-04-04 02:09:57 +08:00

test_responses_circuit_breaker.py

fix: normalize responses circuit breaker keys

2026-04-19 20:16:25 +08:00

test_stepfun_reasoning.py

fix(provider): gate reasoning-to-content fallback behind spec flag

2026-04-26 20:11:08 +08:00

test_transcription.py

fix(transcription): address review nits on PR #3253

2026-05-06 15:52:29 +08:00

test_xiaomi_mimo_thinking.py

fix(providers): wire MiMo thinking control on gateway providers (#3845)

2026-05-16 20:46:34 +08:00