Thiên Toán 9fbd6e619d fix: correct token extraction for Claude non-streaming responses (#131) ...

- Add response logging for non-streaming requests (5_res_provider.json, 7_res_client.json)
- Fix extractUsageFromResponse() to check Claude format before OpenAI format
- Prevents format misidentification that caused tokens to show as 0
- Claude uses input_tokens/output_tokens vs OpenAI's prompt_tokens/completion_tokens

Fixes dashboard Details tab showing 0 tokens for Claude requests

2026-02-20 14:24:21 +07:00

..

config

feat: add Qwen3.5 Coder Model configuration (#156)

2026-02-19 21:55:11 +07:00

executors

feat(responses): respect client streaming preference + string input support (#121)

2026-02-15 11:47:55 +07:00

handlers

fix: correct token extraction for Claude non-streaming responses (#131)

2026-02-20 14:24:21 +07:00

services

feat(responses): respect client streaming preference + string input support (#121)

2026-02-15 11:47:55 +07:00

transformer

feat: enhance request handling and error management in chatCore and streamToJsonConverter

2026-02-15 12:02:53 +07:00

translator

feat(responses): respect client streaming preference + string input support (#121)

2026-02-15 11:47:55 +07:00

utils

fix(open-sse): emit [DONE] in passthrough SSE mode (#142)

2026-02-18 13:26:23 +07:00

.npmignore

Update jsconfig.json and package.json to correct open-sse path references from relative to local directory.

2026-01-05 10:37:09 +07:00

index.js

feat(executors): Improved UI components for displaying provider limits and usage statistics in the dashboard.

2026-02-05 18:38:50 +07:00