Providers

197

clusters

2710

machines

6024

GPUs

191

independent providers

cluster	hardware	model	ttft	single	throughput↓	cache hit
3×	8x NVIDIA B300 SXM6 AC	kimi-k3	0.86s	116 tps	667 tps	84%
1×	8x NVIDIA B300 SXM6 AC	kimi-k3	1.42s	73 tps	58 tps	31%
1×	32x RTX 5090 (32GB)	glm-5.2	0.57s	5 tps	58 tps	51%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	28 tps	38%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	25 tps	53%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	23 tps	—
12×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	22 tps	23%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	20 tps	41%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	18 tps	40%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	17 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	15 tps	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	15 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	2.97s	177 tps	15 tps	88%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	15 tps	25%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	14 tps	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	14 tps	33%
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	13 tps	25%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	13 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	12 tps	24%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	12 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	12 tps	74%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	12 tps	24%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	12 tps	66%
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	12 tps	27%
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	11 tps	24%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	11 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	11 tps	23%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	11 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	11 tps	23%
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	11 tps	23%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	0.94s	194 tps	11 tps	89%
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	11 tps	26%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	10 tps	23%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	10 tps	26%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	10 tps	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	9 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	9 tps	25%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	9 tps	77%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	9 tps	23%
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	9 tps	83%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	9 tps	23%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	9 tps	22%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	9 tps	23%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	9 tps	24%
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	8 tps	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	8 tps	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	7 tps	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	7 tps	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	6 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	6 tps	25%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	5 tps	75%
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	5 tps	73%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	3 tps	85%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	3 tps	94%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	2 tps	—
32×	8x NVIDIA GeForce RTX 4090	qwen3.6-35b-a3b	—	—	2 tps	61%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	2 tps	88%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	1 tps	100%
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	1 tps	83%
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	0 tps	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	0 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	0 tps	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	0 tps	—
5×	8x RTX 5090 (32GB)	qwen3.8-27b	—	—	—	—
11×	32x RTX 5090 (32GB) +1 configs	glm-5.2	—	—	—	—
12×	8x RTX 5090 (32GB)	deepseek-v4-flash-0731	—	—	—	—
1×	8x NVIDIA GeForce RTX 5090	kimi-k3	—	—	—	—
1×	8x4090	qwen3.6-35b-a3b	—	—	—	—
1×	1x NVIDIA RTX 6000 Ada Generation	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
1×	1x NVIDIA RTX PRO 6000 Blackwell Server Edition	qwen3.6-35b-a3b	—	—	—	—
4×	1x NVIDIA L40	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
15×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
1×	4x NVIDIA GeForce RTX 5090	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
1×	2x NVIDIA GeForce RTX 4090	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
32×	8x NVIDIA GeForce RTX 4090	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
1×	1x NVIDIA RTX PRO 6000 Blackwell Server Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
32×	8x NVIDIA GeForce RTX 4090	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
1×	7x NVIDIA L40S	qwen3.6-35b-a3b	—	—	—	—
1×	8x H200	glm-5.2	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
3×	1x NVIDIA GeForce RTX 4090	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
1×	1x NVIDIA RTX 6000 Ada Generation	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
1×	?x GPU	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
64×	8x NVIDIA GeForce RTX 4090	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
13×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
8×	8x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
1×	8x NVIDIA GeForce RTX 5090	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
1×	2x NVIDIA GeForce RTX 5090	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA RTX PRO 6000 Blackwell Workstation Edition	qwen3.6-35b-a3b	—	—	—	—
1×	2x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—
16×	1x NVIDIA H100 80GB HBM3	qwen3.6-35b-a3b	—	—	—	—