Inference provider

Phala

Headquarters: US

20

models

20

endpoints

Phala website preview
Phala website

Catalogue

Available models

ModelEndpointContextInput /MOutput /MCached input /MCache write /MUptime 1d
DeepSeek V3.2phala164K$1.00$1.00$0.50—99.54%
DeepSeek V4 Flash 0731phala1.0M$0.31$0.92$0.02—100.00%
DeepSeek V4 Pro 0813phala1.0M$0.96$2.88$0.10—99.90%
DeepSeek V4.1 Flashphala1.0M$0.21$0.84$0.0042—99.86%
GLM 5.1phala203K$1.21$4.20$0.60—43.75%
GLM 5.2phala/fp81.0M$1.26$3.00$0.22—0.00%
GLM 5.3phala1.0M$0.84$2.64$0.16—99.76%
GLM 5.3 Flashphala/nvfp41.0M$0.11$0.38$0.02—100.00%
Hy3phala262K$0.15$0.64$0.04—99.78%
Kev 4Bphala8K$0.04$0.00——99.87%
Kimi K2.6phala262K$1.09$4.60$0.37—95.91%
Kimi K2.7 Codephala262K$0.95$4.00$0.19—98.69%
Kimi K3phala1.0M$2.55$12.75$0.26—99.79%
Qwen2.5 7B Instructphala33K$0.10$0.20——100.00%
Qwen3.5 397B A17Bphala262K$0.55$3.50$0.22—98.36%
Qwen3.5-27Bphala262K$0.30$2.40$0.03—97.79%
Qwen3.6 27Bphala262K$0.32$3.25$0.03—92.87%
Qwen3.6 35B A3Bphala262K$0.20$1.27$0.06—96.83%
Qwen3.8 27Bphala1M$0.20$2.50$0.05—97.92%
gpt-oss-120bphala131K$0.15$0.60——99.73%