serves endpoints · 2026-07-23

WandB

Official site · 🇺🇸 United States

Models it serves

12 endpoints, cheapest first. This is the view the model-by-model tables cannot give you: what one provider actually offers, and at what precision.

ModelQuant$/M in$/M outContextUptime 30m
gpt-oss-20bfp4$0.03$0.13131,07299.8%
Granite 4.1 8Bbf16$0.05$0.10131,07299.9%
gpt-oss-120bfp4$0.04$0.14131,07299.9%
DeepSeek V4 Flashfp8$0.14$0.281,048,57699.3%
Gemma 4 31Bbf16$0.12$0.35262,14499.9%
Qwen3.6 35B A3Bfp8$0.25$1.25262,144100.0%
Qwen3.5-35B-A3Bfp8$0.25$1.25262,144100.0%
Qwen3.6 27Bfp8$0.60$3.60262,144100.0%
Kimi K2.7 Codeint4$0.94$4.00262,144100.0%
Kimi K2.6fp4$0.95$4.00262,14497.2%
GLM-5.2fp4$1.39$4.40262,14499.9%
DeepSeek V4 Profp8$1.74$3.481,048,57699.6%
◆ Some links on this page are referral links: if you sign up through them this site may earn a commission, at no extra cost to you. Rankings, prices and measurements are taken from the sources listed under Method, and are not influenced by whether a provider has a referral programme.