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

Phala

Official site · origin not confirmed

Models it serves

15 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-20bunknown$0.04$0.15131,07296.6%
Gemma 4 31Bunknown$0.15$0.46262,14494.8%
Gemma 3 27Bunknown$0.15$0.46262,14489.2%
DeepSeek V4 Flashunknown$0.20$0.401,048,57699.3%
gpt-oss-120bunknown$0.15$0.60131,072100.0%
Qwen3.6 35B A3Bunknown$0.20$1.27262,14499.9%
Muse Glimmerunknown$0.35$1.50131,07297.2%
Qwen3.6 27Bunknown$0.32$2.70262,14495.5%
DeepSeek V3.2unknown$1.00$1.00163,840100.0%
Kimi K2.5unknown$0.60$3.00262,144
Qwen3.5 397B A17Bunknown$0.55$3.50262,144100.0%
GLM-5.2fp8$1.13$3.001,048,57699.7%
GLM-5.1unknown$1.21$4.20202,752
Kimi K2.6unknown$1.09$4.60262,14496.3%
Kimi K3unknown$3.00$15.001,048,57699.4%
◆ 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.