DeepSeek R1 Distill Qwen 32B (Q5_K_M) — 23.4 GBon NVIDIA RTX 4090
Overview
DeepSeek R1 Distill Qwen 32B is a 32.76B parameter dense language model by DeepSeek, with code, multilingual, thinking, tool-calls capabilities. It supports a context window of up to 131,072 tokens.
DeepSeek R1 Distill Qwen 32B is a 32.76-billion-parameter dense transformer from DeepSeek, distilled from the larger R1 reasoning model into a Qwen-based architecture. It excels at chain-of-thought reasoning, code generation, and multilingual tasks with built-in thinking capabilities. Compared to standard 30B-class instruct models, it provides stronger logical and mathematical reasoning. The model supports nine languages and a 128K context window, making it suitable for developers and researchers who need reasoning-focused inference on mid-range GPU setups.
At Q5_K_M quantization (medium quality tier), the model weighs 21.66 GB. This fits within the 24 GB of VRAM on NVIDIA RTX 4090, enabling full GPU inference.
The NVIDIA GeForce RTX 4090 is a consumer GPU with 24 GB of GDDR6X VRAM and 1008 GB/s memory bandwidth. It delivers 82.6 FP16 TFLOPS, making it one of the fastest consumer cards for local LLM inference. It handles quantized models up to 20B parameters comfortably. Ideal for home lab builders and developers who want high-throughput inference without datacenter hardware.
Hardware Requirements
| Model size | 21.66 GB |
| VRAM available | 24 GB |
| VRAM used | 23.4 GB |
| System RAM | |
| Min RAM required | 0 GB |
| GPU layers | 64 / 64 |
| Context size | 3,245 |
| Backend | cuda13 |
| Flash attention | Yes |
| Reading from disk | Yes |
Performance Notes
Deploy
Prerequisites
Ensure your GPU nodes are prepared with the NVIDIA container toolkit:
ansible-playbook prositronic.infra.nvidia_container_toolkitCommand
helmfile --state-values-file <(curl -s https://www.prositronic.eu/values/deepseek-r1-distill-qwen-32b/q5_k_m/nvidia-rtx4090.yaml) applyGenerated values.yaml
/values/deepseek-r1-distill-qwen-32b/q5_k_m/nvidia-rtx4090.yaml
Loading values…Frequently Asked Questions
How much VRAM does DeepSeek R1 Distill Qwen 32B (Q5_K_M) need?
The Q5_K_M quantization of DeepSeek R1 Distill Qwen 32B requires 21.66 GB. All 64 layers fit in the 24 GB of VRAM available on NVIDIA RTX 4090, enabling full GPU acceleration.
Can I run DeepSeek R1 Distill Qwen 32B on NVIDIA RTX 4090?
Yes. NVIDIA RTX 4090 provides 24 GB of VRAM, which is enough to run DeepSeek R1 Distill Qwen 32B (Q5_K_M) with all layers on the GPU for optimal performance.
What is quantization?
Quantization reduces a model's numerical precision from its original floating-point format to a more compact representation. This shrinks the file size and VRAM footprint, making it possible to run large models on consumer hardware. The trade-off is a small reduction in output quality. Q5_K_M compresses DeepSeek R1 Distill Qwen 32B from its original size down to 21.66 GB.
What quantization should I choose for DeepSeek R1 Distill Qwen 32B?
Q5_K_M is a medium-quality quantization. Higher-quality quants (Q8, Q6) preserve more model accuracy but need more VRAM. Lower quants (Q4, Q3, Q2) reduce VRAM usage at the cost of some quality. Choose based on your available hardware and quality requirements.
What is flash attention and why is it enabled?
Flash attention is a memory-efficient algorithm that speeds up the attention mechanism in transformer models. It reduces VRAM usage during inference by avoiding the materialisation of the full attention matrix. For DeepSeek R1 Distill Qwen 32B on NVIDIA RTX 4090, flash attention is enabled to maximise context length and throughput within the available 24 GB of VRAM.
How do I run DeepSeek R1 Distill Qwen 32B (Q5_K_M) with Ollama?
Run ollama run deepseek-r1:32b-qwen-distill-q5_k_m to start DeepSeek R1 Distill Qwen 32B (Q5_K_M). Ollama handles downloading the model weights automatically on first run.