Model Spend Arena2ND ED.
serves endpoints · 2026-08-14

BaseTen

Official site · 🇺🇸 United States

Models it serves

9 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
DeepSeek V4 Flashfp8$0.13$0.261,048,57699.7%
gpt-oss-120bfp4$0.10$0.50128,072100.0%
Nemotron 3 Ultra 550Bfp4$0.60$2.40202,80098.8%
Kimi K2.6fp4$0.95$4.00262,000
Inklingfp8$1.00$4.051,048,57699.8%
DeepSeek V4 Profp4$1.32$3.961,048,576100.0%
GLM-5.2fp8$1.40$4.401,048,576100.0%
GLM-5.2fp8$2.10$6.601,048,576100.0%
Kimi K3fp8$3.00$15.001,048,57699.0%
◆ 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.