Can Mac Pro M2 Ultra 64 GB run DeepSeek-V4.1-Flash (748B-A16B)?
64 GB unified · 800 GB/s memory bandwidth · DeepSeek-V4.1-Flash (748B-A16B) has 748B parameters (16B active per token).
Short answer
No. It doesn't fit in memory.
Needs ≈397 GB but only ≈54 GB is available after the OS, other apps and Open WebUI.
Why it doesn’t fit
Even its smallest practical version (4-bit) needs about 397 GB for the model, its context memory and the runtime. After macOS (≈6.0 GB), a browser with a few tabs (≈3.0 GB) and Open WebUI (≈1.0 GB), Mac Pro M2 Ultra 64 GB has about 54 GB left, so it’s 343 GB short.
Ways to make it work
Close everything else
Not enough on its own: with every other app closed it’s still about 340 GB short.
Pick a smaller model from the same family
No smaller DeepSeek-V4 model in our catalog runs well here. See the other models below.
Get more memory
You’d want at least 407 GB of memory in total. The machines below are the cheapest that run it comfortably.
Check your exact setup
These pages assume typical settings. A shorter context or a lighter chat app can change the answer.
Check Mac Pro M2 Ultra 64 GB
The same applies to every workload we rate (chat, coding questions, coding in a repository, agentic coding and long documents): they all need at least as much memory as chat.
Other models for Mac Pro M2 Ultra 64 GB
The best-rated models for chat on this machine.
- Qwen3.6 35B-A3B100–130 tok/sExcellent
- Qwen3-Next 80B-A3B Instruct140–190 tok/sExcellent
- Gemma 4 26B-A4B80–110 tok/sExcellent
- Qwen3.5 35B-A3B100–130 tok/sExcellent
- LFM2.5 24B-A2B190–260 tok/sExcellent
- gpt-oss-20b120–160 tok/sExcellent
Cheapest machines for DeepSeek-V4.1-Flash (748B-A16B)
Hardware rated comfortable or better for chat, lowest price first.
Different tool, context size or workload?
These pages use typical settings. The full check models your exact setup.