DeepSeek R1 Distill 32B
By DeepSeek · China
Updated 2026-07-13
Overview
The 32B DeepSeek R1 distill — the best accessible open-weight reasoner we've tested. Explicit chain-of-thought, MIT-licensed, runs on a single 24GB GPU.
When to pick this model
- Math, logic, and proof-style problems
- Code debugging where explicit reasoning helps
- Research workflows needing visible chain-of-thought
- Self-hosted alternatives to o1-mini-class APIs
- Commercial use under MIT license
VRAM requirements by quantization
| Quantization | VRAM required |
|---|---|
| Q4_K_M (recommended) | 19 GB |
| Q5_K_M | 23 GB |
| Q8_0 | 35 GB |
| FP16 (no quantization) | 64 GB |
VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.
In practice, DeepSeek R1 Distill 32B wants a 24 GB card at Q4_K_M (19 GB). Stepping up to Q8_0 nearly doubles the footprint to 35 GB, and unquantized FP16 weights take 64 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.
Without a GPU, DeepSeek R1 Distill 32B needs roughly 32 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 3 tokens/sec on entry-level GPUs, on the order of 12 tokens/sec on a mid-range card, and up to 30 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.
What hardware do you need
The table below matches DeepSeek R1 Distill 32B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.
| GPU memory | Example cards | Best fit for DeepSeek R1 Distill 32B |
|---|---|---|
| 8 GB | RTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GB | Does not fit — needs 19 GB at Q4_K_M |
| 12 GB | RTX 5070, RTX 5070 Ti Laptop, RTX 4080 Laptop | Does not fit — needs 19 GB at Q4_K_M |
| 16 GB | RTX 5080, RTX 4080 Super, Radeon RX 9070 XT | Does not fit — needs 19 GB at Q4_K_M |
| 24 GB | RTX 4090, Radeon RX 7900 XTX, RTX 5090 Laptop | Q5_K_M (23 GB used) |
| 32 GB | RTX 5090 | Q5_K_M (23 GB used) |
Which GPU should you buy to run DeepSeek R1 Distill 32B?
To run DeepSeek R1 Distill 32B locally at Q4, you need ~19 GB of VRAM. The best value for this is a RTX 4090 (24 GB VRAM).
As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.
Published benchmark scores
| Benchmark | Score |
|---|---|
| AIME 2024 | 72.6 |
| MATH-500 | 94.3 |
| GPQA | 62.1 |
Scores published by the model author or aggregated from public leaderboards. Re-measured monthly by our editorial team.
To put DeepSeek R1 Distill 32B in context: its AIME 2024 score of 72.6 ranks #4 of the 8 catalog models with a published AIME 2024 result (catalog median 71.7); its MATH-500 score of 94.3 ranks #3 of the 8 catalog models with a published MATH-500 result (catalog median 93.3). Scores are author-published and measured under different harnesses, so treat gaps of a few points as noise rather than a verdict.
Strengths
- Best open-weight reasoner that fits on one consumer GPU
- Excellent math and science performance
- Explicit step-by-step thinking
- MIT license
- 32k context
Limitations
- Heavy thinking-token output inflates latency and cost
- Slow time-to-first-useful-answer
- 32k context is shorter than most 2025 peers
- Overkill for simple chat
Typical workloads
In our catalog grid, DeepSeek R1 Distill 32B is filed under Advanced Math, Science, Agents — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: multi-step reasoning and math-flavoured tasks.
The 32k-token context window covers long chats and mid-sized documents, though very large retrieval workloads will need chunking. The MIT license is permissive, so shipping it inside a commercial product raises no special legal questions.
Architecture & training
Architecture: DeepSeek R1 distillation · reinforced chain-of-thought
Training: Distilled from R1 671B · RL on reasoning problems.
The go-to local reasoning model for STEM and code — accept the verbosity, get the accuracy.
Quick start
ollama run deepseek-r1:32bOr use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.
Similar models worth comparing
Frequently asked questions
How much VRAM does DeepSeek R1 Distill 32B need?
At the recommended Q4_K_M quantization, DeepSeek R1 Distill 32B needs about 19 GB of VRAM. Q8_0 takes 35 GB, and unquantized FP16 weights take 64 GB.
Can DeepSeek R1 Distill 32B run without a GPU?
Yes — with roughly 32 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.
What context window does DeepSeek R1 Distill 32B support?
DeepSeek R1 Distill 32B supports a 32k-token context window (32,768 tokens).
Can I use DeepSeek R1 Distill 32B commercially?
Yes. DeepSeek R1 Distill 32B is released under MIT, a permissive open-source license that allows commercial use, modification and redistribution.
How fast is DeepSeek R1 Distill 32B on consumer hardware?
Our compatibility engine estimates on the order of 12 tokens/sec on a mid-range GPU and up to 30 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.
Which quantization of DeepSeek R1 Distill 32B should I download first?
Start with Q4_K_M (19 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. On a 24 GB card you can run up to Q5_K_M.