AI models on fictio
Every AI model on fictio is an open-source model served by Chutes, the decentralized, serverless inference platform that created fictio. fictio offers the full live Chutes catalog of language models, so new open models arrive shortly after release, and you can switch models in the middle of a chat.
Every model listed here runs in a trusted execution environment (TEE) on Chutes.
Text models
14 language models are available right now. The context window is how much of the conversation a model can read at once.
| Model | Organization | Context window | Capabilities | On Chutes |
|---|---|---|---|---|
DeepSeek-V3.2deepseek-ai/DeepSeek-V3.2-TEE | deepseek-ai | 128K tokens | TEE | View DeepSeek-V3.2 on Chutes |
DeepSeek-V4-Flash-0731deepseek-ai/DeepSeek-V4-Flash-0731-TEE | deepseek-ai | 1M tokens | TEE | View DeepSeek-V4-Flash-0731 on Chutes |
gemma-4-31B-turbogoogle/gemma-4-31B-turbo-TEE | 128K tokens | Image inputTEE | View gemma-4-31B-turbo on Chutes | |
GLM-5.1zai-org/GLM-5.1-TEE | zai-org | 198K tokens | TEE | View GLM-5.1 on Chutes |
GLM-5.2zai-org/GLM-5.2-TEE | zai-org | 1M tokens | TEE | View GLM-5.2 on Chutes |
Kimi-K2.6moonshotai/Kimi-K2.6-TEE | moonshotai | 256K tokens | Image inputTEE | View Kimi-K2.6 on Chutes |
Kimi-K3moonshotai/Kimi-K3-TEE | moonshotai | 1M tokens | Image inputTEE | View Kimi-K3 on Chutes |
Mistral-Nemo-Instruct-2407unsloth/Mistral-Nemo-Instruct-2407-TEE | unsloth | 128K tokens | TEE | View Mistral-Nemo-Instruct-2407 on Chutes |
Nemotron-3-Nano-Omni-30BNemotron-3-Nano-Omni-30B-TEE | — | 128K tokens | TEE | View Nemotron-3-Nano-Omni-30B on Chutes |
Qwen3-32BQwen/Qwen3-32B-TEE | Qwen | 40K tokens | TEE | View Qwen3-32B on Chutes |
Qwen3-235B-A22B-Thinking-2507Qwen/Qwen3-235B-A22B-Thinking-2507-TEE | Qwen | 256K tokens | TEE | View Qwen3-235B-A22B-Thinking-2507 on Chutes |
Qwen3.5-397B-A17BQwen/Qwen3.5-397B-A17B-TEE | Qwen | 256K tokens | Image inputTEE | View Qwen3.5-397B-A17B on Chutes |
Qwen3.6-27BQwen/Qwen3.6-27B-TEE | Qwen | 256K tokens | Image inputTEE | View Qwen3.6-27B on Chutes |
Qwen3.8-27BQwen/Qwen3.8-27B-TEE | Qwen | 256K tokens | Image inputTEE | View Qwen3.8-27B on Chutes |
Images, video and voice
Studio image generation (text-to-image and image-to-image), image-to-video, and text-to-speech on fictio also run on Chutes. Models currently used include:
- Text-to-imagez-image-turbo on Chutes
- Text-to-imageQwen-Image-2512 on Chutes
- Image-to-imageQwen-Image-Edit-2511 on Chutes
- Image-to-videoturbowani2v (Wan 2.2) on Chutes
- Text-to-speechKokoro on Chutes