Models

GLM 5.3 Flash Uncensored. Lower token prices than GLM 5.3, with browser chat and API access.

GLM 5.3 Uncensored, available through browser chat and the same Chat Completions API. Compare it on your own tasks before choosing.

MiMo V2.6 Flash Uncensored for chat, coding and tool workflows. A published 1M-token context window and separate cached-input pricing.

DeepSeek V4.1 Flash 1017 Uncensored with reasoning, streaming and function tool calling. A published 1M-token context window, up to 65,536 output tokens and separate cached-input pricing.

Qwen 3.8 27B on our own hardware, used for the 10 free replies. API and credit-funded Chat can use a compatible backup when the primary is unavailable.