deepseek-r1:8b

Family context: DeepSeek-R1 distillation releases

Publisher: DeepSeek AI

DeepSeek released smaller dense models trained with reasoning samples from DeepSeek-R1. These distills use different Qwen or Llama upstream bases and are distinct from the full R1 mixture-of-experts release.

Publisher-described context, separate from Sigma5C observations and recorded catalog statistics.

Published family information

1.5B and 7B distill bases

Qwen2.5-Math-1.5B and Qwen2.5-Math-7B Source

8B distill base

Llama-3.1-8B Source

Publisher-described uses

• Reasoning-oriented text generation using the documented distilled checkpoints. Source

Reference limitations

• The full R1 model’s 671B architecture and context table do not describe a smaller distill. The source also identifies separate upstream licenses for Qwen- and Llama-derived releases. Source

A recorded tag and an upstream card association do not verify the deployed checkpoint, quantization, configured context or current hosted availability.

Publisher sources

• DeepSeek-R1 distillation releases official documentation — DeepSeek AI. Checked 2026-11-02.

Links

Recorded references do not establish downloadable weights, licensing or current availability.

• DeepSeek — official site
• Ollama
• Hugging Face
• Official model announcement / documentation
• GitHub (maker)

Where deepseek-r1:8b comes from

Named in 1 observation page

• 3B-7B local models — all 53 · V1