---
id: "inference-net"
site: "inference.net"
headquarters: "US"
gateways: ["openrouter", "vercel"]
models: 12
endpoints: 38
tokens_week: 449050235747
token_share: "2.7%"
rank: 10
cheapest_on: "11 of 38 shared models"
cheaper_than: "83% of offers at the median"
vs_card: "31% off the card price at the median over 19 models"
cache_hit: "91.2%"
uptime_1d: 99.7000
latency_p50_ms: 497
tokens_per_second: 84
trains_on_some: true
retains_on_some: true
quantization: "fp4 (12), bf16 (6), fp8 (4), mxfp4 (4), unknown (3)"
---
# InferenceNet

`deepseek/deepseek-v4.1-flash`, `deepseek/deepseek-v4.1-flash-1017`, `inference-net/schematron-v2-small`, `inference-net/schematron-v2-turbo`, `minimax/minimax-m3.1-flash`, `moonshotai/kimi-k3`, `moonshotai/kimi-k3.1-pro`, `xiaomi/mimo-v2.6-flash`, `z-ai/glm-5.3`, `zai/glm-5.3`, `z-ai/glm-5.3-flash`, `zai/glm-5.3-flash`