Spaces:

---

umint

/

ai

like 167

Running

Today a new model has been added, DeepSeek V3.1, which is the latest release from DeepSeek. The reasoning mode is off by default, and to enable it the setting "reasoning_effort" must be adjusted in Chat Controls.

Several previous models have also been restored, including:

• DeepSeek: R1 (0528)
• Qwen: Qwen 3 (235B-A22B)
• Qwen: Qwen 3 Coder (480B-A35B)

Disabled model:

• OpenAI: GPT-OSS (20B)

Like this project? Feel free to buy me a coffee.