Will It Run AI

Can DeepSeek R1 Distill 32B run on RTX 5090 32GB?

YES — Tight Fit

A78Great
Estimated from fit model

DeepSeek R1 Distill 32B needs ~27.8 GB VRAM. RTX 5090 32GB has 32.0 GB. With Q4_K_M quantization, expect ~66 tok/s.

Runtime: OllamaCapacity: TightBandwidth: HighStack: BasicBottleneck: Balanced
Share:

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 27.8 GB, 66.4 tok/s, Tight fit
27.8 GB required32.0 GB available
87% VRAM used

Fit status

Tight fit

Decode

66.4 tok/s

TTFT

2914 ms

Safe context

33K

Memory

27.8 GB / 32.0 GB

Memory breakdown

Weights19.5 GB
KV Cache3.9 GB
Runtime1.2 GB
Headroom3.2 GB

See how fast it feels

See how fast it feelsDeepSeek R1 Distill 32B on RTX 5090 32GB
1st promptCold start — includes initialization
>Solve: If a train travels at 60 mph for 2.5 hours, how far does it go?
Let me work through this step by step. Given information: - Speed = 60 mph (miles per hour) - Time = 2.5 hours Formula: Distance = Speed x Time Calculation: Distance = 60 mph x 2.5 h = 150 miles The train travels 150 miles. Double-check: at 60 mph the train covers 60 miles each hour. In half an hour it covers 30 miles. So in 2.5 hours: 60 + 60 + 30 = 150 miles. Confirmed.
2nd promptWarm — model ready, no init overhead
>What if the train stops for 20 minutes halfway?
The stop doesn't change the total distance covered while moving, but it affects total trip time. Breakdown: - First half: 1.25 hours at 60 mph = 75 miles - Stop: 20 minutes (no distance) - Second half: 1.25 hours at 60 mph = 75 miles Total distance: 150 miles (same) Total time: 2.5 h + 0.33 h = 2.83 hours Average speed: 150 / 2.83 = 53 mph The distance stays the same but average speed drops to 53 mph because of the stop.
Estimated: 66.4 tok/s decode · 2.9s TTFT (warm) · 166 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

No major red flags

This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.

Best improvement path

Performance by workload

WorkloadGradeFitDecodeTTFTContext
ChatARuns well66.4 tok/s1590 ms33K
CodingATight fit66.4 tok/s2914 ms33K
Agentic CodingARuns with offload66.4 tok/s4239 ms33K
ReasoningATight fit66.4 tok/s3444 ms33K
RAGARuns with offload66.4 tok/s5299 ms33K

Quantization options

How DeepSeek R1 Distill 32B (32B params) fits at each quantization level on RTX 5090 32GB (32.0 GB usable).

QuantBitsVRAMQualityFit
Q2_K
2
12.5 GB
LowA73
Q3_K_S
3
15.7 GB
LowA75
NVFP4
4
17.9 GB
MediumA75
Q4_K_M
4
19.5 GB
MediumA74
Q5_K_MBest for your GPU
5
23.0 GB
HighA74
Q6_K
6
26.2 GB
HighF0
Q8_0
8
34.2 GB
Very HighF0
F16
16
65.6 GB
MaximumF0

Get started

Copy-paste commands to run DeepSeek R1 Distill 32B on your machine.

Run

ollama run deepseek-r1:32b

Your hardware

More models your RTX 5090 32GB can run

ModelParamsGradeDecodeCapabilities
AlibabaQwen 3.6 35B A3B35BS128.2 tok/s
AlibabaQwen 3.5 35B A3B35BS139.4 tok/s
Moonshot AIKimi Linear 48B A3B48BA26.7 tok/s
IBMGranite Code 34B34BA62.7 tok/s
CohereCommand R 35B35BA61.2 tok/s

Frequently asked questions

Can RTX 5090 32GB run DeepSeek R1 Distill 32B?

Yes, RTX 5090 32GB can run DeepSeek R1 Distill 32B with a A grade (Tight fit). Expected decode speed: 66.4 tok/s.

How much VRAM does DeepSeek R1 Distill 32B need?

DeepSeek R1 Distill 32B (32B parameters) requires approximately 27.8 GB of memory with Q4_K_M quantization.

What is the best quantization for DeepSeek R1 Distill 32B?

The recommended quantization for DeepSeek R1 Distill 32B is Q4_K_M, which balances quality and memory efficiency.

What speed will DeepSeek R1 Distill 32B run at on RTX 5090 32GB?

On RTX 5090 32GB, DeepSeek R1 Distill 32B achieves approximately 66.4 tokens per second decode speed with a time-to-first-token of 2914ms using Q4_K_M quantization.

Can RTX 5090 32GB run DeepSeek R1 Distill 32B for coding?

For coding workloads, DeepSeek R1 Distill 32B on RTX 5090 32GB receives a A grade with 66.4 tok/s and 33K context.

What context window can DeepSeek R1 Distill 32B use on RTX 5090 32GB?

On RTX 5090 32GB, DeepSeek R1 Distill 32B can safely use up to 33K tokens of context. The model's official context limit is 33K, but available memory constrains the safe maximum.

See all results for RTX 5090 32GBSee all hardware for DeepSeek R1 Distill 32B
Embed this result▼

Paste this snippet into any page to show a live fit card.

<iframe src="https://willitrunai.com/embed/deepseek-r1-distill-32b-on-rtx-5090-32gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: