Inference.net
Inference platform and maker of the Schematron extraction models
inference.netSan Francisco, United States
Inference.net is a San Francisco company, started in 2023 as the distributed GPU network Kuzco before rebranding. It offers hosted open models, tracing, evaluation and fine-tuning, and trains Schematron models for converting HTML into structured JSON.
- Founded
- 2023
- Type
- Private company
- Founders
- Sam Hogan and Ibrahim Ahmed
- AI lab
- Inference provider
- 22 hosted endpoints
| Model | Endpoint | Context | Input /M | Output /M | Cached input /M | Cache write /M | Uptime 1d |
|---|---|---|---|---|---|---|---|
 DeepSeek V4.1 Flash | inference-net/fp8 | 1M | $0.103 | $0.60 | $0.008 | — | 100.00% |
| DeepSeek V4.1 Flash | inference-net/us | 1M | $0.103 | $0.60 | $0.008 | — | 100.00% |
| DeepSeek V4.1 Flash 1017 | inference-net/fp8 | 1M | $0.08 | $0.605 | $0.008 | — | 100.00% |
| DeepSeek V4.1 Flash 1017 | inference-net/us | 1M | $0.08 | $0.605 | $0.008 | — | 100.00% |
 GLM 5.3 | inference-net/fp4 | 1M | $0.08 | $3.60 | $0.076 | — | 100.00% |
| GLM 5.3 | inference-net/us | 1M | $0.08 | $3.60 | $0.076 | — | 100.00% |
| GLM 5.3 Flash | inference-net/fp4 | 1M | $0.108 | $0.35 | $0.035 | — | 100.00% |
| GLM 5.3 Flash | inference-net/us | 1M | $0.108 | $0.35 | $0.035 | — | 100.00% |
 Kimi K3 | inference-net/fast | 1M | $0.45 | $18.00 | $0.315 | — | 99.93% |
| Kimi K3 | inference-net/fp4 | 1M | $0.234 | $15.00 | $0.204 | — | 100.00% |
| Kimi K3 | inference-net/us | 1M | $0.234 | $15.00 | $0.204 | — | 99.93% |
| Kimi K3 | inference-net/us/fast | 1M | $0.45 | $18.00 | $0.315 | — | 100.00% |
| Kimi K3.1 Pro | inference-net/fast | 1M | $2.70 | $13.50 | $0.27 | — | 100.00% |
| Kimi K3.1 Pro | inference-net/mxfp4 | 1M | $1.80 | $9.00 | $0.18 | — | 100.00% |
| Kimi K3.1 Pro | inference-net/mxfp4/us | 1M | $1.80 | $9.00 | $0.18 | — | 100.00% |
| Kimi K3.1 Pro | inference-net/us/fast | 1M | $2.70 | $13.50 | $0.27 | — | 100.00% |
 MiniMax M3.1 Flash | inference-net/fp4 | 1M | $0.15 | $0.60 | $0.015 | — | 100.00% |
| MiniMax M3.1 Flash | inference-net/fp4/us | 1M | $0.15 | $0.60 | $0.015 | — | 99.61% |
| Schematron V2 Small | inference-net/bf16 | 128K | $0.05 | $0.23 | $0.05 | — | 100.00% |
| Schematron V2 Small | inference-net/us | 128K | $0.05 | $0.23 | $0.05 | — | 100.00% |
| Schematron V2 Turbo | inference-net/bf16 | 128K | $0.03 | $0.15 | $0.03 | — | 100.00% |
| Schematron V2 Turbo | inference-net/us | 128K | $0.03 | $0.15 | $0.03 | — | 100.00% |