Chutes
Decentralized confidential inference on Bittensor with TEE
Chutes is a decentralized AI compute marketplace built on the Bittensor network. TEE-tagged models run inside AMD SEV-SNP and Intel TDX enclaves, providing verifiable privacy on community-run hardware. Offers the widest selection of open-source models with hardware-attested confidentiality.
API Features
OpenAI Compatible ✓
Streaming ✓
Function Calling —
Batch Inference —
Vision —
Embeddings —
Models & Pricing
| Model | Family | Context | Max Out | Input /M | Output /M | Free |
|---|---|---|---|---|---|---|
| Qwen3 32B | Qwen | 40K | 40K | $0.08 | $0.24 | — |
| Gemma 4 31B | Google Gemma | 128K | 64K | $0.13 | $0.38 | — |
| DeepSeek V3.1 | DeepSeek | 160K | 64K | $0.27 | $1.00 | — |
| Kimi K2.5 | Moonshot AI | 256K | 64K | $0.44 | $2.00 | — |
| DeepSeek V3.2 | DeepSeek | 128K | 64K | $0.28 | $0.42 | — |
| GPT-OSS 120B | OpenAI GPT | 128K | 64K | $0.09 | $0.36 | — |
| MiMo V2 Flash | MiMo | 256K | 64K | $0.09 | $0.29 | — |
| GLM 5.1 | ZhipuAI GLM | 198K | 64K | $1.05 | $3.50 | — |
| Qwen3.5 397B A17B | Qwen | 256K | 64K | $0.39 | $2.34 | — |
| DeepSeek V3 0324 | DeepSeek | 160K | 64K | $0.25 | $1.00 | — |
| MiniMax M2.5 | MiniMax | 192K | 64K | $0.15 | $1.20 | — |
| Qwen3 235B A22B | Qwen | 256K | 64K | $0.10 | $0.60 | — |
| GLM 4.7 | ZhipuAI GLM | 198K | 64K | $0.39 | $1.75 | — |
| DeepSeek R1T2 Chimera | DeepSeek | 160K | 160K | $0.30 | $1.10 | — |
| GLM 5 | ZhipuAI GLM | 198K | 64K | $0.95 | $2.55 | — |
| DeepSeek R1 0528 | DeepSeek | 160K | 64K | $0.45 | $2.15 | — |
| Kimi K2.6 | Moonshot AI | 256K | 64K | $0.95 | $4.00 | — |
| Qwen3-Coder Next | Qwen | 256K | 64K | $0.12 | $0.75 | — |
| Qwen3.6 27B | Qwen | 256K | 64K | $0.20 | $1.56 | — |