Model · DeepSeek-R1 · Reviewed June 2026

standardMIT14B paramsOpen weights

DeepSeek-R1 Distill 14B

DeepSeek · DeepSeek-R1

14B parameter open-weight model. Qwen2.5 distillation of R1. Strong reasoning and coding at 14B. Q4 tight in 12 GB; comfortable in 16 GB VRAM.

Editorial review

Reviewed byOpenSourcesAI EditorialLast updatedJune 2026SourcesHuggingFace model card (deepseek-ai/DeepSeek-R1-Distill-Qwen-14B), official docs, OpenSourcesAI editorial review.

VRAM figures are empirical estimates. Actual usage varies by runtime, context length, and system configuration. Verify on your specific hardware before production use.

Related models

DeepSeek-R1 Distill 8B8B parameter model.DeepSeek-V4-ProTeams comparing frontier-style open-weight reasoning and coding models against hosted closed models.DeepSeek R1Reasoning experiments, coding workflows, and evaluating open reasoning models against closed alternatives.

Ready to run this model locally?

Find a compatible interface in our Local AI Tools directory →