CLIProxyAPI

fix(gemini): remove default thinking for gemini 3 models

hkfires · 2026-01-02 10:55:59 +08:00

fdf5720217

feat(thinking): add numeric budget to thinkingLevel conversion fallback

hkfires · 2025-12-31 17:14:47 +08:00

d00e3ea973

fix(thinking): use model alias for thinking config resolution in mapped models

hkfires · 2025-12-31 17:09:22 +08:00

89db4e9481

feat(amp): enhance model mapping and Gemini thinking configuration

This commit introduces several improvements to the AMP (Advanced Model Proxy) module:

- **Model Mapping Logic:** The `FallbackHandler` now uses a more robust approach for model mapping. It includes the extraction and preservation of dynamic "thinking suffixes" (e.g., `(xhigh)`) during mapping, ensuring that these configurations are correctly applied to the mapped model. A new `resolveMappedModel` function centralizes this logic for cleaner code.
- **ModelMapper Verification:** The `ModelMapper` in `model_mapping.go` now verifies that the target model of a mapping has available providers *after* normalizing it. This prevents mappings to non-existent or unresolvable models.
- **Gemini Thinking Configuration Cleanup:** In `gemini_thinking.go`, unnecessary `generationConfig.thinkingConfig.include_thoughts` and `generationConfig.thinkingConfig.thinkingBudget` fields are now deleted from the request body when applying Gemini thinking levels. This prevents potential conflicts or redundant configurations.
- **Testing:** A new test case `TestModelMapper_MapModel_TargetWithThinkingSuffix` has been added to `model_mapping_test.go` to specifically cover the preservation of thinking suffixes during model mapping.

Luis Pater · 2025-12-20 22:19:35 +08:00

ed5ec5b55c

fix(gemini): add optional skip for gemini3 thinking conversion

hkfires · 2025-12-19 22:07:43 +08:00

2039062845

fix(util): disable default thinking for gemini 3 flash

hkfires · 2025-12-19 13:11:15 +08:00

13aa82f3f3

fix: require dot in gemini25Pattern regex for precise matching

Ben Vargas · 2025-12-17 16:09:50 -07:00

88798816f2

fix: apply thinkingLevel from model suffix metadata for Gemini 3

The previous commit added thinkingLevel support but didn't apply it
when the reasoning effort came from model name suffix (e.g., model(minimal)).

This was because ResolveThinkingConfigFromMetadata returns nil for
level-based models, bypassing the metadata application.

Changes:
- Add ApplyGemini3ThinkingLevelFromMetadata for standard Gemini API
- Add ApplyGemini3ThinkingLevelFromMetadataCLI for CLI API format
- Update gemini_cli_executor to apply Gemini 3 thinkingLevel from metadata
- Update antigravity_executor to apply Gemini 3 thinkingLevel from metadata
- Update aistudio_executor to apply Gemini 3 thinkingLevel from metadata
- Add comprehensive test coverage for Gemini 3 thinkingLevel functions

Ben Vargas · 2025-12-17 16:08:38 -07:00

598f0af19b

feat: use thinkingLevel for Gemini 3 models per Google documentation

Per Google's official documentation, Gemini 3 models should use
thinkingLevel (string) instead of thinkingBudget (number) for
optimal performance.

From Google's Gemini Thinking docs:
> Use the thinkingLevel parameter with Gemini 3 models. While
> thinkingBudget is accepted for backwards compatibility, using
> it with Gemini 3 Pro may result in suboptimal performance.

Changes:
- Add model family detection functions (IsGemini3Model, IsGemini25Model,
  IsGemini3ProModel, IsGemini3FlashModel)
- Add ApplyGeminiThinkingLevel and ApplyGeminiCLIThinkingLevel functions
  for applying thinkingLevel config
- Add ValidateGemini3ThinkingLevel for model-specific level validation
- Add ThinkingBudgetToGemini3Level for backward compatibility conversion
- Update NormalizeGeminiThinkingBudget to convert budget to level for
  Gemini 3 models
- Update ApplyDefaultThinkingIfNeeded to not set a default level for
  Gemini 3 (lets API use its dynamic default "high")
- Update ConvertThinkingLevelToBudget to preserve thinkingLevel for
  Gemini 3 models
- Add Levels field to all Gemini 3 model definitions:
  - Gemini 3 Pro: ["low", "high"]
  - Gemini 3 Flash: ["minimal", "low", "medium", "high"]

Backward compatibility:
- Gemini 2.5 models continue to use thinkingBudget as before
- If thinkingBudget is provided for Gemini 3, it's converted to the
  appropriate thinkingLevel
- Existing configurations continue to work

Ben Vargas · 2025-12-17 15:28:20 -07:00

a33f5d31fc

fix(thinking): normalize effort mapping

Route OpenAI reasoning effort through ThinkingEffortToBudget for Claude
translators, preserve "minimal" when translating OpenAI Responses, and
treat blank/unknown efforts as no-ops for Gemini thinking configs.

Also map budget -1 to "auto" and expand cross-protocol thinking tests.

hkfires · 2025-12-15 09:16:15 +08:00

d20b71deb9

fix(thinking): centralize reasoning_effort mapping

Move OpenAI `reasoning_effort` -> Gemini `thinkingConfig` budget logic into
shared helpers used by Gemini, Gemini CLI, and antigravity translators.

Normalize Claude thinking handling by preferring positive budgets, applying
budget token normalization, and gating by model support.

Always convert Gemini `thinkingBudget` back to OpenAI `reasoning_effort` to
support allowCompat models, and update tests for normalization behavior.

hkfires · 2025-12-15 09:16:14 +08:00

716aa71f6e

fix(thinking): map budgets to effort levels

Ensure thinking settings translate correctly across providers:
- Only apply reasoning_effort to level-based models and derive it from numeric
  budget suffixes when present
- Strip effort string fields for budget-based models and skip Claude/Gemini
  budget resolution for level-based or unsupported models
- Default Gemini include_thoughts when a nonzero budget override is set
- Add cross-protocol conversion and budget range tests

hkfires · 2025-12-12 21:33:20 +08:00

374faa2640

feat(util): implement dynamic thinking suffix normalization and refactor budget resolution logic

- Added support for parsing and normalizing dynamic thinking model suffixes.
- Centralized budget resolution across executors and payload helpers.
- Retired legacy Gemini-specific thinking handlers in favor of unified logic.
- Updated executors to use metadata-based thinking configuration.
- Added `ResolveOriginalModel` utility for resolving normalized upstream models using request metadata.
- Updated executors (Gemini, Codex, iFlow, OpenAI, Qwen) to incorporate upstream model resolution and substitute model values in payloads and request URLs.
- Ensured fallbacks handle cases with missing or malformed metadata to derive models robustly.
- Refactored upstream model resolution to dynamically incorporate metadata for selecting and normalizing models.
- Improved handling of thinking configurations and model overrides in executors.
- Removed hardcoded thinking model entries and migrated logic to metadata-based resolution.
- Updated payload mutations to always include the resolved model.

Luis Pater · 2025-12-11 03:10:50 +08:00

423ce97665

fix(executor): centralize default thinking config

hkfires · 2025-12-09 21:05:06 +08:00

9b202b6c1c

feat(executor): normalize thinking budget across all Gemini executors

hkfires · 2025-12-09 21:05:06 +08:00

5ec9b5e5a9

**feat(util): add -reasoning suffix support for Gemini models**

Adds support for the `-reasoning` model name suffix which enables
thinking/reasoning mode with dynamic budget. This allows clients to
request reasoning-enabled inference using model names like
`gemini-2.5-flash-reasoning` without explicit configuration.

The suffix is normalized to the base model (e.g., gemini-2.5-flash)
with thinkingBudget=-1 (dynamic) and include_thoughts=true.

Follows the existing pattern established by -nothinking and
-thinking-N suffixes.

auroraflux · 2025-11-30 01:18:57 -08:00

32d3809f8c

Fixed: #291

**feat(executor): add thinking level to budget conversion utility**

- Introduced `ConvertThinkingLevelToBudget` to map thinking level ("high"/"low") to corresponding budget values.
- Applied the utility in `aistudio_executor.go` before stripping unsupported configs.
- Updated dependencies to include `tidwall/gjson` for JSON parsing.

Luis Pater · 2025-11-21 00:48:12 +08:00

cbcfeb92cc

Add AMP fallback proxy and shared Gemini normalization

- add fallback handler that forwards Amp provider requests to ampcode.com when the provider isn’t configured locally
- wrap AMP provider routes with the fallback so requests always have a handler
- share Gemini thinking model normalization helper between core handlers and AMP fallback

Ben Vargas · 2025-11-19 18:23:17 -07:00

8193392bfe

fix(executor): only apply thinking config to supported models

hkfires · 2025-10-29 19:19:17 +08:00

7dd93a4a25

Feature: #103

feat(gemini): add Gemini thinking configuration support and metadata normalization

- Introduced logic to parse and apply `thinkingBudget` and `include_thoughts` configurations from metadata.
- Enhanced request handling to include normalized Gemini model metadata, preserving the original model identifier.
- Updated Gemini and Gemini-CLI executors to apply thinking configuration based on metadata overrides.
- Refactored handlers to support metadata extraction and cloning during request preparation.

Luis Pater · 2025-10-16 11:31:18 +08:00

ade279d1f2

20 Commits