Releases · JIAQIA/whosellm · GitHub

Repository navigation

Release list

v0.3.1 Latest

Latest

JIAQIA released this 28 Sep 15:27

e593162

Full Changelog: v0.3.0...v0.3.1

Assets 2

v0.3.0

JIAQIA released this 08 Sep 05:03

109fb4c

Full Changelog: v0.2.7...v0.3.0

Assets 2

v0.2.7

JIAQIA released this 04 Sep 07:14

b9d0063

Full Changelog: v0.2.6...v0.2.7

Assets 2

v0.2.6

JIAQIA released this 04 Sep 04:23

7a10820

Full Changelog: v0.2.5...v0.2.6

Assets 2

v0.2.5

JIAQIA released this 20 Aug 06:54

8915b01

Full Changelog: v0.2.4...v0.2.5

Assets 2

v0.2.4

JIAQIA released this 10 Jun 03:27

0be7f5b

Full Changelog: v0.2.3...v0.2.4

Assets 2

v0.2.3

JIAQIA released this 08 Jun 08:27

049a7c5

Full Changelog: v0.2.2...v0.2.3

Assets 2

v0.2.2

JIAQIA released this 24 Apr 04:26

3ce0dfc

Full Changelog: v0.2.1...v0.2.2

Assets 2

v0.2.1

JIAQIA released this 12 Apr 12:58

18ead5c

Full Changelog: v0.2.0...v0.2.1

Assets 2

v0.2.0

JIAQIA released this 12 Apr 08:50

3fb9e8c

What's New

E2E 元数据测试框架

新增官方文档驱动的 E2E 测试，覆盖 4 个供应商 37 个模型，验证 LLMeta 解析结果与官方文档一致。

OpenAI: 11 models (GPT-5.4, GPT-4o, O4)
Anthropic: 8 models (Claude 4.6/4.5/4.1/4.0/3.0)
Google: 9 models (Gemini 3.x/2.5/2.0)
ZhipuAI: 9 models (GLM-5/4.7/4.6/4.5, GLM-Vision)

模型配置修复

基于官方文档核实，修复多项模型元数据：

Anthropic: Claude Opus/Sonnet 4.6 context window 升级至 1M tokens，补全 computer_use 支持
OpenAI: GPT-5.4-pro structured_outputs 修正，O4 系列 thinking 支持修正
ZhipuAI: GLM-4.5 系列 structured_outputs 修正，GLM-4.5V max_tokens 修正

Full Changelog: v0.1.1a5...v0.2.0

Assets 2