Skip to content
Docs.Albastov.

Chat models

Open-weight Qwen and DeepSeek chat models on one OpenAI-compatible route.

All 5

Qwen3.8 27Bqwen3.8-27b

The default chat model, for general chat, tools and JSON output with 150,000 tokens of context.

Chat

More models 4

Qwen3.8 27B Fastqwen3.8-27b-fast

Qwen3.8 27B tuned for faster answers to a few requests at once.

Chat

Qwen3.8 27B Uncensoredqwen3.8-27b-uncensored

Qwen3.8 27B with refusals removed and shorter reasoning, 150,000 tokens of context.

Chat

Qwen3.8 Flash-Next Abliteratedqwen3.8-flash-next-abliterated

Qwen3.8 Flash-Next, a mixture-of-experts model with refusals removed, one request at a time.

Chat

DeepSeek V4.1 Flashdeepseek-v4.1-flash

DeepSeek V4.1 Flash, a large mixture-of-experts model for short prompts, one request at a time.

Chat