Model Spend Arena2ND ED.
serves endpoints · 2026-10-03

InferenceNet

Official site · origin not confirmed

Models it serves

7 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
MiMo-V2.6-Flashfp8$0.12$0.281,048,57699.5%
DeepSeek V4.1 Flashunknown$0.04$0.601,040,00099.8%
GLM-5.3-Flashfp4$0.05$0.601,048,57694.3%
GLM-5.2unknown$0.10$2.201,048,576100.0%
GLM-5.3unknown$0.22$3.391,048,576100.0%
Kimi K3fp4$0.99$13.001,048,576100.0%
Kimi K3fp4$3.00$15.00250,000100.0%