Dayuan Jiang 72d647de7a fix: use Chat Completions API for OpenAI-compatible proxies (#382) ...

Third-party OpenAI-compatible proxies typically don't support the
/responses endpoint. Use .chat() for custom baseURLs while keeping
Responses API for official OpenAI to preserve reasoning model support.

Fixes #377

2025-12-23 20:29:48 +09:00

..

i18n

feat(i18n): add translations for model configuration UI (#368)

2025-12-23 11:42:27 +09:00

types

feat: multi-provider model configuration with UI/UX improvements (#355)

2025-12-22 22:36:36 +09:00

ai-config.ts

refactor: eliminate code duplication (DRY principle) (#211)

2025-12-11 14:28:02 +09:00

ai-providers.ts

fix: use Chat Completions API for OpenAI-compatible proxies (#382)

2025-12-23 20:29:48 +09:00

base-path.ts

Support subdirectory deployment and fix API path handling (#311)

2025-12-22 23:28:55 +09:00

cached-responses.ts

refactor: simplify LLM XML format to output bare mxCells only (#254)

2025-12-14 14:04:44 +09:00

dynamo-quota-manager.ts

feat: add server-side quota tracking with DynamoDB (#379)

2025-12-23 18:36:27 +09:00

langfuse.ts

refactor: simplify Langfuse integration with AI SDK 6 (#375)

2025-12-23 16:26:45 +09:00

pdf-utils.ts

feat: make PDF/text extraction char limit configurable via env (#214)

2025-12-11 14:14:31 +09:00

storage.ts

feat: multi-provider model configuration with UI/UX improvements (#355)

2025-12-22 22:36:36 +09:00

system-prompts.ts

feat: add get_shape_library tool for AI icon discovery (#335)

2025-12-20 23:19:49 +09:00

token-counter.ts

feat: add daily token limit with actual usage tracking (#171)

2025-12-08 18:56:34 +09:00

use-file-processor.tsx

refactor: eliminate code duplication (DRY principle) (#211)

2025-12-11 14:28:02 +09:00

use-quota-manager.tsx

feat: add server-side quota tracking with DynamoDB (#379)

2025-12-23 18:36:27 +09:00

utils.ts

fix: enable progressive diagram rendering during streaming (#380)

2025-12-23 18:54:03 +09:00