1059 Commits

• Fix Gemini function-calling INVALID_ARGUMENT by relaxing Gemini tool validation and cleaning schema
  sowar1987 · 2026-01-22 17:11:07 +08:00
  a2f8f59192
• refactor(translator): enhance signature handling in Claude and Gemini requests, streamline cache usage and remove unnecessary tests
  hkfires · 2026-01-21 20:21:49 +08:00
  c8884f5e25
• refactor(cache, translator): refine signature caching logic and tests, replace session-based logic with model group handling
  Luis Pater · 2026-01-21 18:30:05 +08:00
  d9c6317c84
• refactor(cache, translator): remove session ID from signature caching and clean up logic
  Luis Pater · 2026-01-21 13:37:10 +08:00
  ef4508dbc8
• refactor(translator): remove session ID logic from signature caching and associated tests
  Luis Pater · 2026-01-21 12:45:07 +08:00
  f775e46fe2
• refactor(cache): simplify signature caching by removing sessionID parameter
  Luis Pater · 2026-01-21 12:38:05 +08:00
  65ad5c0c9d
• fix(translator): update HasValidSignature to require modelName parameter for improved validation
  Luis Pater · 2026-01-21 11:31:37 +08:00
  88bf4e77ec
• test(logging): add unit tests for GinLogrusRecovery middleware panic handling
  Luis Pater · 2026-01-21 10:57:27 +08:00
  a4f8015caa
• fix(translator): preserve thinking blocks by skipping signature
  hkfires · 2026-01-21 10:49:20 +08:00
  9332316383
• fix(executor): only strip maxOutputTokens for non-claude models
  hkfires · 2026-01-21 10:49:20 +08:00
  6dcbbf64c3
• feat(cache): handle gemini family in signature cache with fallback validator logic
  Luis Pater · 2026-01-21 10:11:21 +08:00
  2ce3553612
• feat(translator): enhance ConvertGeminiRequestToAntigravity with model name and refine reasoning block handling
  Luis Pater · 2026-01-21 08:31:23 +08:00
  2e14f787d4
• test(responses): add comprehensive tests for SSE event ordering and response transformations
  Luis Pater · 2026-01-21 07:08:59 +08:00
  523b41ccd2
• Merge pull request #1117 from router-for-me/cache

  fix(translator): enhance signature cache clearing logic and update test cases with model name

  Luis Pater · 2026-01-20 23:18:48 +08:00
  c6fa1d0e67
• Merge pull request #1116 from bexcodex/fix/antigravity

  Fix antigravity malformed_function_call

  Luis Pater · 2026-01-20 22:40:00 +08:00
  ac56e1e88b
• fix(translator): enhance signature cache clearing logic and update test cases with model name
  hkfires · 2026-01-20 20:02:29 +08:00
  9b72ea9efa
• Fix antigravity malformed_function_call
  bexcodex · 2026-01-20 19:54:54 +08:00
  9f364441e8
• chore(translator): update cache functions to include model name parameter in tests
  Luis Pater · 2026-01-20 18:36:51 +08:00
  e49a1c07bf
• feat(translator): unify model group references by introducing GetModelGroup helper function
  Luis Pater · 2026-01-20 13:45:25 +08:00
  8d9f4edf9b
• feat(translator): improve signature handling by associating with model name in cache functions
  Luis Pater · 2026-01-20 13:31:36 +08:00
  020e61d0da
• Fixed: #1109

  feat(translator): enhance session ID derivation with user_id parsing in Claude

  Luis Pater · 2026-01-20 12:35:40 +08:00
  6184c43319
• chore(translator): remove unnecessary whitespace in gemini_openai_response code
  Luis Pater · 2026-01-20 11:47:33 +08:00
  2cbe4a790c
• Merge branch 'main' into dev (PR #961)
  Luis Pater · 2026-01-20 11:42:22 +08:00
  68b3565d7b
• feat(auth): hash account ID for improved uniqueness in credential filenames
  Luis Pater · 2026-01-20 11:37:52 +08:00
  9823dc35e1
• feat(auth): add hashed account ID to credential filenames for team plans
  Luis Pater · 2026-01-20 11:36:29 +08:00
  059bfee91b
• Merge pull request #877 from zhiqing0205/main

  feat(codex): include plan type in auth filename

  Luis Pater · 2026-01-20 11:11:25 +08:00
  1fef90ff58
• fix(login): remove emojis from interactive prompt messages
  Luis Pater · 2026-01-20 11:09:56 +08:00
  8447fd27a0
• refactor(claude): remove redundant system instructions check in Claude executor
  Luis Pater · 2026-01-20 11:02:52 +08:00
  7831cba9f6
• Merge pull request #868
  Luis Pater · 2026-01-20 10:57:24 +08:00
  e02b2d58d5
• Merge pull request #861 from umairimtiaz9/fix/gemini-cli-backend-project-id

  fix(auth): use backend project ID for free tier Gemini CLI OAuth users

  Luis Pater · 2026-01-20 10:32:17 +08:00
  28726632a9
• Merge pull request #1108 from router-for-me/modelinfo

  feat(registry): support provider-specific model info lookup

  Luis Pater · 2026-01-20 10:18:42 +08:00
  3b26129c82
• refactor(antigravity): remove unused client signature handling in thinking objects
  Luis Pater · 2026-01-20 10:17:55 +08:00
  d4bb4e6624
• Merge branch 'dev' into fix/cross-model-thinking-signature
  Luis Pater · 2026-01-20 10:10:43 +08:00
  a7ffc77e3d
• feat(registry): support provider-specific model info lookup
  hkfires · 2026-01-20 10:01:17 +08:00
  e641fde25c
• feat(vertex): add Imagen image generation model support

  Add support for Imagen 3.0 and 4.0 image generation models in Vertex AI:
  
  - Add 5 Imagen model definitions (4.0, 4.0-ultra, 4.0-fast, 3.0, 3.0-fast)
  - Implement :predict action routing for Imagen models
  - Convert Imagen request/response format to match Gemini structure like gemini-3-pro-image
  - Transform prompts to Imagen's instances/parameters format
  - Convert base64 image responses to Gemini-compatible inline data

  dinhkarate · 2026-01-20 01:26:37 +07:00
  8734d4cb90
• Merge pull request #1099 from router-for-me/claude

  refactor(claude): move max_tokens constraint enforcement to Apply method

  Luis Pater · 2026-01-19 20:55:59 +08:00
  5baa753539
• fix(executor): stop rewriting thinkingLevel for gemini
  hkfires · 2026-01-19 19:49:39 +08:00
  1d2fe55310
• feat(registry): expand antigravity model config

  Remove static Name mapping and add entries for claude-sonnet-4-5,
  tab_flash_lite_preview, and gpt-oss-120b-medium configs

  hkfires · 2026-01-19 19:32:00 +08:00
  c175821cc4
• feat(claude): clamp thinking budget to max_tokens constraints
  hkfires · 2026-01-19 16:32:20 +08:00
  239a28793c
• refactor(claude): move max_tokens constraint enforcement to Apply method
  hkfires · 2026-01-19 15:50:35 +08:00
  c421d653e7
• fix(translator): avoid forcing RFC 8259 system prompt
  hkfires · 2026-01-19 11:33:27 +08:00
  52e46ced1b
• feat(translator): report cached token usage in Claude output
  hkfires · 2026-01-19 11:23:44 +08:00
  cf9daf470c
• feat(translator): add reasoning state tracking and improve reasoning summary handling

  - Introduced `oaiToResponsesStateReasoning` to track reasoning data.
  - Enhanced logic for emitting reasoning summary events and managing state transitions.
  - Updated output generation to handle multiple reasoning entries consistently.

  Luis Pater · 2026-01-19 03:58:28 +08:00
  140d6211cc
• fix(thinking): update ValidateConfig to include fromSuffix parameter and adjust budget validation logic
  hkfires · 2026-01-18 16:37:14 +08:00
  cb6caf3f87
• Merge pull request #1067 from router-for-me/auth-files

  refactor(auth): simplify filename prefixes for qwen and iflow tokens

  Luis Pater · 2026-01-18 13:41:59 +08:00
  99c7abbbf1
• Merge pull request #1076 from sususu98/fix/antigravity-enum-string

  fix(antigravity): convert non-string enum values to strings for Gemini API

  Luis Pater · 2026-01-18 13:40:53 +08:00
  8f511ac33c
• Merge pull request #1081 from router-for-me/thinking

  Refine thinking validation and cross‑provider payload conversion

  Luis Pater · 2026-01-18 13:34:28 +08:00
  f88228f1c5
• refactor(logging): centralize log directory resolution logic

  - Introduced `ResolveLogDirectory` function in `logging` package to standardize log directory determination across components.
  - Replaced redundant logic in `server`, `global_logger`, and `handlers` with the new utility function.

  Luis Pater · 2026-01-18 12:40:57 +08:00
  62e2b672d9
• refactor(thinking): add Gemini family provider grouping for strict validation
  hkfires · 2026-01-18 11:30:53 +08:00
  03005b5d29
• refactor(thinking): pass source and target formats to ApplyThinking for cross-format validation

  Update ApplyThinking signature to accept fromFormat and toFormat parameters
  instead of a single provider string. This enables:
  
  - Proper level-to-budget conversion when source is level-based (openai/codex)
    and target is budget-based (gemini/claude)
  - Strict budget range validation when source and target formats match
  - Level clamping to nearest supported level for cross-format requests
  - Format alias resolution in SDK translator registry for codex/openai-response
  
  Also adds ErrBudgetOutOfRange error code and improves iflow config extraction
  to fall back to openai format when iflow-specific config is not present.

  hkfires · 2026-01-18 10:30:15 +08:00
  c7e8830a56