YUNWU API

A unified API gateway providing access to multiple large language models with direct connectivity in China.

Categories

YUNWU API offers 21 LLM API models.

Speed benchmark average: 134 tok/s.

YUNWU API is an API aggregator, offering models from multiple vendors.

Avg Speed132.45 tok/s

Latency6.02 s

Total Tests135

Models21

Updated4/12/2026

Created At8/13/2025

Supported Models

Model	Speed	Latency	Tests
mistral-small-latest	474.47 tok/s	1.51s	5
gemini-2.5-flash-lite-preview-06-17	381.52 tok/s	0.84s	5
gemini-2.5-flash-lite-nothinking	318.15 tok/s	0.99s	10
gpt-5.4-nano	274.17 tok/s	1.68s	5
gpt-5.4-mini	200.70 tok/s	1.44s	5
gpt-5-nano	153.35 tok/s	9.46s	15
gpt-5-nano-2025-08-07	111.99 tok/s	10.26s	5
grok-4-1-fast-non-reasoning	111.23 tok/s	3.38s	5
gemini-2.5-pro	101.10 tok/s	17.38s	5
gemini-2.5-pro-thinking	95.90 tok/s	4.10s	5
gemini-2.5-pro-exp-03-25	93.19 tok/s	16.05s	5
grok-4-fast	84.15 tok/s	4.01s	5
gpt-4o-mini	83.79 tok/s	8.66s	5
mimo-v2-flash	60.50 tok/s	2.09s	10
qwen3-coder-480b-a35b-instruct	58.38 tok/s	1.43s	10
gpt-4o	57.59 tok/s	2.58s	5
qwq-plus-latest	43.91 tok/s	19.49s	5
claude-sonnet-4-5-20250929	37.74 tok/s	4.77s	5
deepseek-reasoner	26.79 tok/s	23.92s	5
deepseek-v3-0324	22.64 tok/s	1.52s	10

Showing 20 of 21 models.