CLIProxyAPI

refactor(logging): streamline auth selection debug messages

Reduce duplicate Debugf calls by appending proxy info via an optional suffix and keep the debug-level guard inside the helper.

Michael Velbaum · 2025-12-28 16:10:11 +02:00

cb3bdffb43

refactor(logging): pass request entry into auth selection log

Avoid re-creating the request-scoped log entry in the helper and use a switch for account type dispatch.

Michael Velbaum · 2025-12-28 15:51:11 +02:00

48f19aab51

refactor(logging): dedupe auth selection debug logs

Extract repeated debug logging for selected auth credentials into a helper so execute, count, and stream paths stay consistent.

Michael Velbaum · 2025-12-28 15:42:35 +02:00

48f6d7abdf

fix(logging): quote OAuth account field

Use strconv.Quote when embedding the OAuth account in debug logs so unexpected characters (e.g. quotes) can't break key=value parsing.

Michael Velbaum · 2025-12-28 15:32:54 +02:00

79fbcb3ec4

feat(logging): disambiguate OAuth credential selection in debug logs

When multiple OAuth providers share an account email, the existing "Use OAuth" debug lines are ambiguous and hard to correlate with management usage stats. Include provider, auth file, and auth index in the selection log, and only compute these fields when debug logging is enabled to avoid impacting normal request performance.

Before:
[debug] Use OAuth user@example.com for model gemini-3-flash-preview
[debug] Use OAuth user@example.com (project-1234) for model gemini-3-flash-preview

After:
[debug] Use OAuth provider=antigravity auth_file=antigravity-user_example_com.json auth_index=1a2b3c4d5e6f7788 account="user@example.com" for model gemini-3-flash-preview
[debug] Use OAuth provider=gemini-cli auth_file=gemini-user@example.com-project-1234.json auth_index=99aabbccddeeff00 account="user@example.com (project-1234)" for model gemini-3-flash-preview

Michael Velbaum · 2025-12-28 15:22:36 +02:00

0e4148b229

Merge pull request #757 from ben-vargas/fix-thinking-toolchoice-conflict

Fix: disable thinking when tool_choice forces tool use

Luis Pater · 2025-12-28 14:04:30 +08:00

v6.6.63 457924828a

Fix: disable thinking when tool_choice forces tool use

Anthropic API does not allow extended thinking when tool_choice is set
to "any" or a specific tool. This was causing 400 errors when using
features like Amp's /handoff command which forces tool_choice.

Added disableThinkingIfToolChoiceForced() that removes thinking config
when incompatible tool_choice is detected, applied to both streaming
and non-streaming paths.

Fixes router-for-me/CLIProxyAPI#630

Ben Vargas · 2025-12-27 16:31:37 -07:00

aca2ef6359

feat(management): add generic API call handler to management endpoints

Luis Pater · 2025-12-28 04:40:32 +08:00

ade7194792

feat(cliproxy): implement model aliasing and hashing for Codex configurations, enhance request routing logic, and normalize Codex model entries

Luis Pater · 2025-12-28 03:06:51 +08:00

v6.6.62 3a436e116a

Merge pull request #756 from leaph/check-ai-thinking-settings

feat(iflow): add model-specific thinking configs for GLM-4.7 and Mini…

Luis Pater · 2025-12-28 02:08:27 +08:00

v6.6.61 336867853b

feat(iflow): add model-specific thinking configs for GLM-4.7 and MiniMax-M2.1

- GLM-4.7: Uses extra_body={"thinking": {"type": "enabled"}, "clear_thinking": false}
- MiniMax-M2.1: Uses reasoning_split=true for OpenAI-style reasoning separation
- Added preserveReasoningContentInMessages() to support re-injection of reasoning
  content in assistant message history for multi-turn conversations
- Added ThinkingSupport to MiniMax-M2.1 model definition

leaph · 2025-12-27 18:39:15 +01:00

6403ff4ec4

Update issue templates

Luis Pater · 2025-12-28 01:22:42 +08:00

d222469b44

Fixed: #749

fix(translators): ensure `gjson.String` content is non-empty before setting `parts` in OpenAI request logic

Luis Pater · 2025-12-28 00:54:26 +08:00

v6.6.60 7646a2b877

Merge pull request #750 from router-for-me/config

fix(config): preserve original config structure and avoid default value pollution

Luis Pater · 2025-12-27 22:10:01 +08:00

62090f2568

Fixed: #747

fix(translators): rename and integrate `usageMetadata` as `cpaUsageMetadata` in Claude processing logic

Luis Pater · 2025-12-27 22:02:11 +08:00

v6.6.59 c281f4cbaf

fix(config): make streaming keepalive and retries ints

hkfires · 2025-12-27 20:56:47 +08:00

09455f9e85

fix(config): smart merge writes non-default new keys only

hkfires · 2025-12-27 20:28:54 +08:00

c8e72ba0dc

docs(config): clarify merge mapping behavior

hkfires · 2025-12-27 19:30:21 +08:00

375ef252ab

chore(config): update ignore patterns

hkfires · 2025-12-27 19:13:14 +08:00

ee552f8720

fix(config): avoid adding new keys when merging

hkfires · 2025-12-27 19:00:47 +08:00

2e88c4858e

Merge pull request #745 from router-for-me/auth

fix(auth): make provider rotation atomic

Luis Pater · 2025-12-27 13:01:22 +08:00

3f50da85c1

fix(auth): make provider rotation atomic

hkfires · 2025-12-27 12:56:48 +08:00

8be06255f7

Fixed: #738

fix(translators): refine prompt token calculation by incorporating cached tokens in Claude response handling

Luis Pater · 2025-12-27 03:56:11 +08:00

v6.6.58 72274099aa

Fixed: #736

fix(translators): handle gjson string types in Claude request processing to ensure consistent content parsing

Luis Pater · 2025-12-27 01:25:47 +08:00

v6.6.57 dcae098e23

Merge pull request #727 from nguyenphutrong/main

docs: add Quotio to community projects

Luis Pater · 2025-12-26 11:53:09 +08:00

2eb05ec640

feat(usage): add import/export functionality for usage statistics and enhance deduplication logic

Luis Pater · 2025-12-26 11:49:51 +08:00

v6.6.56 3ce0d76aa4

docs(readme): add Quotio to community projects section

Trong Nguyen · 2025-12-26 10:46:05 +07:00

a00b79d9be

fix(translators): ensure correct handling and output of multimodal assistant content across request handlers

Luis Pater · 2025-12-26 05:08:04 +08:00

v6.6.55 33e53a2a56

Merge pull request #722 from hungthai1401/bugfix/remove-extra-args

Fixed incorrect function signature call to `NewBaseAPIHandlers`

Luis Pater · 2025-12-26 02:56:56 +08:00

cd5b80785f

fix(test): remove extra argument from ExecuteStreamWithAuthManager call

Thai Nguyen Hung · 2025-12-25 21:55:35 +07:00

54f71aa273

Merge pull request #704 from tinyc0der/add-index

fix(openai): add index field to image response for LiteLLM compatibility

Luis Pater · 2025-12-25 21:35:12 +08:00

3f949b7f84

feat(config): add commercial-mode to optimize HTTP middleware for lower memory usage

Luis Pater · 2025-12-25 21:05:01 +08:00

v6.6.54 443c4538bb

refactor(image): avoid using json.Marshal

TinyCoder · 2025-12-25 14:21:01 +07:00

a7fc2ee4cf

docs(readme): update GLM model version from 4.6 to 4.7 in README and README_CN

Luis Pater · 2025-12-24 23:59:48 +08:00

8e749ac22d

docs(readme): update GLM model version from 4.6 to 4.7 in README and README_CN

Luis Pater · 2025-12-24 23:46:27 +08:00

69e09d9bc7

Fixed: #696

fix(translators): adjust prompt token calculation by subtracting cached tokens across Gemini, OpenAI, and Claude handlers

Luis Pater · 2025-12-24 23:29:18 +08:00

v6.6.53 06ad527e8c

Merge pull request #706 from router-for-me/log

Log

Luis Pater · 2025-12-24 22:24:39 +08:00

v6.6.52 b7409dd2de

refactor(logging): standardize request id formatting and layout

hkfires · 2025-12-24 22:03:07 +08:00

5ba325a8fc

Merge pull request #695 from NguyenSiTrung/main

feat: add cached token parsing for Gemini , Antigravity API responses

Luis Pater · 2025-12-24 21:58:55 +08:00

d502840f91

fix(logging): normalize warning level to warn

hkfires · 2025-12-24 21:11:37 +08:00

99238a4b59

refactor(logging): inline request id in log output

hkfires · 2025-12-24 21:07:18 +08:00

6d43a2ff9a

Merge pull request #700 from router-for-me/log

refactor(sdk/auth): rename manager.go to conductor.go

Luis Pater · 2025-12-24 19:36:24 +08:00

v6.6.51 3faa1ca9af

feat(models): add support for GLM-4.7 and MiniMax-M2.1

Luis Pater · 2025-12-24 19:30:57 +08:00

9d975e0375

feat(api): add endpoint to retrieve request logs by ID

hkfires · 2025-12-24 19:24:51 +08:00

2a6d8b78d4

fix(openai): add index field to image response for LiteLLM compatibility

LiteLLM's Pydantic model requires an index field in each image object.
Without it, responses fail validation with "images.0.index Field required".

TinyCoder · 2025-12-24 17:43:31 +07:00

671558a822

refactor(sdk/auth): rename manager.go to conductor.go

hkfires · 2025-12-24 15:21:03 +08:00

26fbb77901

Merge remote-tracking branch 'upstream/main'

NguyenSiTrung · 2025-12-24 10:54:09 +07:00

a277302262

refactor: extract parseGeminiFamilyUsageDetail helper to reduce duplication

NguyenSiTrung · 2025-12-24 10:22:31 +07:00

969c1a5b72

feat: add cached token parsing for Gemini API responses

NguyenSiTrung · 2025-12-24 10:20:11 +07:00

872339bceb

Merge pull request #697 from Cubence-com/main

docs(readme): add Cubence sponsor and fix PackyCode link

Luis Pater · 2025-12-24 11:19:32 +08:00

5dc0dbc7aa

1175 Commits