← All modelsVERIFIED MODEL PROFILE · ESTABLISHED

DeepSeek R1 32B

Distilled reasoning variants for step-by-step analysis and mathematics.

SAFE RAM FLOOR26 GB
MODEL FILE~20 GB
INPUTText
LICENCEMIT / base-model terms

Will it run?

LocalLens uses 26 GB as a conservative starting floor for this Ollama profile. That means the model file, runtime and operating system have some space. It does not promise speed. Your processor, graphics support, context length and other open programs still matter.

Check it against your RAM →

What is it useful for?

ReasoningMathResearch

Verified Ollama setup

Install Ollama from its official website, then run:

ollama run deepseek-r1:32b

Start with a real but non-sensitive task. Check the answer, response time and device temperature before using private data.

Evidence and limits

The published maximum context is 128K published max. Maximum context is not a promise that the full context fits at the minimum RAM floor.

Verify this model on Ollama ↗