text model · DeepSeek-R1-Distill · Windows
Can I run DeepSeek-R1-Distill-Qwen 32B on AMD Radeon RX 7900 XTX (24GB)?
Yes. DeepSeek-R1-Distill-Qwen 32B runs on AMD Radeon RX 7900 XTX (24GB) at Q4_K_M (~22.1 GB of ~23 GB usable).
Fits at Q4_K_M (~22.1 GB of ~23 GB usable) but with little headroom. Close other apps; a smaller context frees a few hundred MB.
That figure is at a 4k context and moves about ±15% as context length changes. AMD Radeon RX 7900 XTX (24GB) leaves ~0.9 GB of headroom.
- Q4_K_M needed
- ~22.1 GB
- Usable on device
- ~23 GB
- Device memory
- 24 GB
- Best quant
- Q4_K_M
Which quant fits
Running cost · estimate
- Power draw
- ~355 W
- Electricity / 1M tokens
- ~$0.48
- Pays for itself after
- ~49,950M tok
At ~$0.15/kWh and the estimated ~31 tok/s, a million generated tokens costs about $0.48 in electricity. Versus a hosted API at ~$0.5 per million tokens, the ~$999 AMD Radeon RX 7900 XTX (24GB) pays for itself after roughly 49,950 million tokens, so local hardware is mostly a fixed cost, not a per-token one. TDP is the peak draw, so this is an upper bound. Assumptions.
Run it
Pick your tool. All 3 load the same Q4_K_M weights.
ollama run deepseek-r1:32b llama-cli -hf bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF:Q4_K_M lms get bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF - Parameters
- 32B
- Q4_K_M size
- 19.85 GB
- Q8_0 size
- 34.82 GB
- Context
- 128k
- Ollama tag
- deepseek-r1:32b
- Memory
- 24 GB vram
- Usable for weights
- ~23 GB
- Power draw
- ~355 W
- Best runtime
- Ollama (ROCm) / llama.cpp ROCm (Linux)
You could also run
Run DeepSeek-R1-Distill-Qwen 32B on other hardware
FAQ
Can AMD Radeon RX 7900 XTX (24GB) run DeepSeek-R1-Distill-Qwen 32B?
Yes. DeepSeek-R1-Distill-Qwen 32B runs on AMD Radeon RX 7900 XTX (24GB) at Q4_K_M (~22.1 GB of ~23 GB usable).
How much memory does DeepSeek-R1-Distill-Qwen 32B need?
It is a tight fit on AMD Radeon RX 7900 XTX (24GB). At Q4_K_M the weights are ~19.85 GB; with KV cache and runtime overhead, budget ~22.1 GB at a 4k context.
What is the best tool to run DeepSeek-R1-Distill-Qwen 32B on Windows?
LM Studio for a simple setup; Ollama (CUDA) for the most speed. AMD GPUs run via Vulkan/ROCm at roughly half CUDA throughput. NVIDIA is the smooth path on Windows.
Embed this
A live badge for your model card or README, updated as the data is. [](https://localmodel.run/can-i-run/deepseek-r1-distill-qwen-32b/amd-rx-7900-xtx-24gb) Sources
Weights are measured from GGUF files; KV cache and overhead are computed, so totals can vary ~15% with context and runtime. Any tok/s is a bandwidth estimate. See methodology.