SShogo Bench

CoreWeave

21 models served.

ModelAgent IndexInputOutputAgent $/MUptime 24hContextQuant
DeepSeek V4.1 Flash 101757.6$0.30$1.20$0.10100.0%1Mfp8
GLM-5.355.2$0.75$3.00$0.4099.7%1Mnvfp4
DeepSeek V4.1 Flash51.2$0.13$0.65$0.050100.0%1Mfp8
GLM-5.3-Flash48.3$0.12$0.38$0.05099.9%1Mnvfp4
GLM-5.247.3$0.62$2.42$0.32100.0%1Mfp4
Kimi K2.643.7$0.65$3.41$0.3899.8%262Kfp4
Kimi K2.7 Code42.1$0.60$3.00$0.3398.5%262Kint4
MiniMax-M336.7$0.23$0.96$0.1296.9%262Kfp4
GPT OSS 120B22.1$0.030$0.17$0.04098.9%131Kfp4
Beam 1.0 501B A23B—$0.30$1.00$0.1297.2%1Mnvfp4
DeepSeek V3.1—$0.50$1.50$0.30100.0%161Kfp8
DeepSeek V4 Flash 0731—$0.13$0.27$0.07099.9%262Kfp8
DeepSeek V4 Pro 0813—$1.31$3.96$0.41100.0%1Mfp8
Gemma 4 26B A4B IT—$0.10$0.30$0.070100.0%262Kbf16
Gemma 4 31B IT—$0.10$0.34$0.11100.0%262Kfp4
GPT OSS 20B—$0.030$0.13$0.03099.6%131Kfp4
Granite 4.2 8B—$0.060$0.15$0.03098.8%131Kbf16
Nemotron 3.5 Lightning 30B A3B—$0.070$0.20$0.050100.0%262Kbf16
Nemotron 3.5 Super VL—$0.15$0.47$0.060100.0%1Mnvfp4
Qwen3.6 35B-A3B—$0.10$0.89$0.090100.0%262Kfp8
Qwen3.8 27B—$0.40$3.00$0.3299.4%262Kfp8