> Beta. Some things may not work as described yet.

# Five models.

GLM, Kimi, DeepSeek and Qwen over one attested API. USD per million tokens, input / output. Every model runs on an attested GPU in sealed hardware. The signed price sheet is at `https://api.sealedin.ai/v1/catalog`.

| Model | Price |
|---|---:|
| GLM-5.2 `chutes/zai-org/GLM-5.2-TEE` | 1.31 / 4.15 |
| Kimi-K3 `chutes/moonshotai/Kimi-K3-TEE` | 1.94 / 9.98 |
| DeepSeek-V4-Flash `chutes/deepseek-ai/DeepSeek-V4-Flash-0731-TEE` | 0.46 / 1.39 |
| Qwen3.5-397B `chutes/Qwen/Qwen3.5-397B-A17B-TEE` | 0.47 / 3.15 |
| Qwen3.8-27B `chutes/Qwen/Qwen3.8-27B-TEE` | 0.041 / 1.31 |

Prices as of 2026-11-03.

[How the GPU attestation is checked](/trust.md)