DeepSeek R1 on M5 Ultra 512 GB: VRAM, speed and fit
DeepSeek R1 at 3-bit runs on the M5 Ultra 512 GB with 160K of context at about 30 tokens a second.
Text generation · Chat · Ollama / LM Studio
Runs comfortably · on M5 Ultra 512 GB · 3-bit (Q3_K_M) · Ollama / LM Studio · Chat
- GPU memory
- 451 GB
- of 512 GB
- Context
- 160K
- auto
- Speed
- 30 tok/s
- 21 when full
- First reply
- 7.8 s
- 2K prompt
Where the memory goes
- Model
- 312 GB
- Context
- 10.7 GB
- Overhead
- 733 MB
- OS
- 128 GB
- Free
- 61.0 GB
- Memory
- 451 GB of 512 GB
- Context
- 160K tokens · auto
- Speed
- 30 tok/s · 21 when full
- First reply
- 7.8 s for a 2K prompt
- Running cost
- $5.06/mo at home · $574/mo on a cloud 8× A40
Validation: Not validated
| Check | Target | Result | Status |
|---|---|---|---|
| Fits in GPU memory | 384 GB usable | 323 GB needed | Pass |
| Headroom | 5% free | 61.0 GB free (16%) | Pass |
| Context for chat | 16K or more | 160K | Pass |
| Writing speed | 15 tok/s or more | 30 tok/s | Pass |
| First reply | under 3 s, 2K prompt | 7.8 s | Fails |
Targets: a reply you can read along with, a first reply the use case tolerates, and 5% of memory left free. Calculated from published specs, not benchmarked.
The setup
- Model
- DeepSeek R1
- Parameters
- 684B, 40.1B active
- Precision
- 3-bit (Q3_K_M)
- Context
- 160K, auto
- Chats at once
- 1
- Use
- Chat
- Runs in
- Ollama / LM Studio
- Hardware
- M5 Ultra 512 GB
- GPU can use
- 384 GB of 512 GB
Running cost
- At home
- $5.06/mo
- 231 W at the wall
- Renting instead
- $574/mo
- 8× A40 at $4.72/h
4 h a day, power at $0.18 per kWh, cloud at RunPod on-demand prices as of Nov 22, 2026.
How to run it
- llama.cpp
llama-server -hf unsloth/DeepSeek-R1-GGUF- vLLM
vllm serve deepseek-ai/DeepSeek-R1- MLX (Mac)
mlx_lm.chat --model mlx-community/DeepSeek-R1-4bit- Model card
- huggingface.co/deepseek-ai/DeepSeek-R1
DeepSeek R1 on similar machines
- M5 Max 128 GBToo big
- M3 Ultra 96 GBToo big
- M3 Ultra 256 GBRuns
- M3 Ultra 512 GBRuns
- Ryzen AI Max+ 395 64 GBToo big
- Ryzen AI Max+ 395 128 GBToo big
- RTX SparkJust misses
- DGX SparkJust misses
All machines: DeepSeek R1 VRAM requirements. Everything that runs here: what can the M5 Ultra 512 GB run?
Other models that run on M5 Ultra 512 GB
Questions
Can DeepSeek R1 run on M5 Ultra 512 GB?
DeepSeek R1 at 3-bit runs on the M5 Ultra 512 GB with 160K of context at about 30 tokens a second.
Report VR-1S4K0JE, calculated Nov 22, 2026 from published specs, not benchmarked.