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

Together

Official site · 🇺🇸 United States

Models it serves

13 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 Flashunknown$0.14$0.281,048,57699.6%
Qwen3.5-9Bunknown$0.17$0.25262,14499.9%
GLM-5.3-Flashunknown$0.15$0.501,048,57599.9%
gpt-oss-120bunknown$0.15$0.60131,072100.0%
MiniMax-M3unknown$0.30$1.20524,28899.4%
DeepSeek V4.1 Flashunknown$0.30$1.201,048,576100.0%
Muse Glimmerunknown$0.35$1.50131,072100.0%
Inklingunknown$1.00$4.05524,288100.0%
DeepSeek V4 Prounknown$1.32$3.961,048,57697.9%
GLM-5.2unknown$1.40$4.401,048,57599.8%
GLM-5.3unknown$1.40$4.401,048,57599.2%
Qwen3.8 2.4T A95Bunknown$2.00$6.001,010,000100.0%
Kimi K3unknown$2.70$13.501,048,57699.9%