联无所AI

联无所AI provides an OpenAI-compatible API relay service for accessing multiple AI models with competitive pricing.

Categories

联无所AI offers 47 LLM API models.

Speed benchmark average: 66 tok/s.

Avg Speed66.14 tok/s

Latency7.93 s

Total Tests278

Models47

Updated4/16/2026

Created At8/13/2025

Supported Models

Model	Speed	Latency	Tests
Qwen/Qwen2-1.5B-Instruct	213.84 tok/s	0.67s	5
Pro/Qwen/Qwen2-1.5B-Instruct	204.04 tok/s	0.60s	5
免费Qwen2-1.5B	203.36 tok/s	0.68s	5
免费Grok3-mini	180.93 tok/s	3.99s	5
deepseek-ai/deepseek-vl2	125.77 tok/s	0.75s	5
Pro/Qwen/Qwen2-7B-Instruct	96.25 tok/s	0.63s	5
deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B	94.25 tok/s	5.51s	5
Qwen/Qwen2-7B-Instruct	93.64 tok/s	0.68s	5
Pro/Qwen/Qwen2-VL-7B-Instruct	93.23 tok/s	0.71s	5
免费Qwen2-7B	93.01 tok/s	0.85s	10
免费DS-VL2	86.80 tok/s	1.86s	5
免费Grok3	85.51 tok/s	1.39s	5
Pro/Qwen/Qwen2.5-VL-7B-Instruct	85.51 tok/s	0.76s	5
免费Qwen2.5-14B	77.95 tok/s	0.69s	5
Qwen/Qwen2.5-14B-Instruct	77.92 tok/s	0.69s	5
免费Qwen2-VL-7B	77.82 tok/s	0.95s	10
internlm/internlm2_5-7b-chat	73.88 tok/s	0.61s	5
免费GLM-4-9B-128K	73.70 tok/s	0.77s	5
Qwen/QwQ-32B-Preview	72.57 tok/s	0.81s	5
沉浸式翻译	72.24 tok/s	1.01s	5

Showing 20 of 47 models.