Memory at each precision
| Version | Weights | With context | Smallest Mac |
|---|---|---|---|
| 8-bit | 302 GB | 326 GB | 512 GB |
| 6-bit | 231 GB | 249 GB | 512 GB |
| 4-bit | 160 GB | 173 GB | 256 GB |
| 3-bit lower quality | 124 GB | 134 GB | 192 GB |
| 2-bit lower quality | 88.8 GB | 95.9 GB | 128 GB |
On every Apple chip
The smallest memory size of each chip that runs DeepSeek V4 Flash 0731, the best version that fits it, and the writing speed you can expect.
| Chip | Needs | Version | Tokens/s |
|---|---|---|---|
| M6 | no size fits | Too big | – |
| M5 Ultra | 256 GB | 4-bit Runs well | 74–132 |
| M5 Max | 128 GB | 2-bit Runs well | 67–120 |
| M5 Pro | no size fits | Too big | – |
| M5 | no size fits | Too big | – |
| M4 Max | 128 GB | 2-bit Runs well | 60–107 |
| M4 Pro | no size fits | Too big | – |
| M4 | no size fits | Too big | – |
| M3 Ultra | 256 GB | 4-bit Runs well | 50–90 |
| M3 Max | 128 GB | 2-bit Runs well | 44–78 |
| M3 Pro | no size fits | Too big | – |
| M3 | no size fits | Too big | – |
| M2 Ultra | 128 GB | 2-bit Runs well | 88–156 |
| M2 Max | no size fits | Too big | – |
| M2 Pro | no size fits | Too big | – |
| M2 | no size fits | Too big | – |
| M1 Ultra | 128 GB | 2-bit Runs well | 88–156 |
| M1 Max | no size fits | Too big | – |
| M1 Pro | no size fits | Too big | – |
| M1 | no size fits | Too big | – |
How to run it
- The easy way: download MLX Harness, free and open source. Search for DeepSeek V4 Flash 0731, download the version that fits your Mac, and chat. It also serves the model to your other apps over an OpenAI-compatible API.Download MLX HarnessFree · Apple Silicon · macOS 14.5 or later
- In Terminal with Apple’s MLX:
pip install mlx-lmmlx_lm.chat --model mlx-community/DeepSeek-V4-Flash-4bitBuild: mlx-community/DeepSeek-V4-Flash-4bit on Hugging Face. Other precisions sit next to it.
Step by step: download and run LLMs on a Mac · how to use MLX Harness · MLX vs Ollama vs llama.cpp
About DeepSeek V4 Flash 0731
| Made by | DeepSeek |
|---|---|
| Parameters | 284B total · 13B active (mixture of experts) |
| Context | 1M tokens |
| Licence | MIT |
| Released | 2026-07 |
| Input | Text |
| Weights | deepseek-ai/DeepSeek-V4-Flash-0731 |
Questions
- How much memory does DeepSeek V4 Flash 0731 need?
- About 160 GB of weights at 4-bit, plus room for the context: 173 GB in all. On a Mac that means 256 GB of unified memory or more, since macOS lets the GPU use about 75% of it by default.
- How fast is DeepSeek V4 Flash 0731 on a Mac?
- Writing speed depends on memory bandwidth. The fastest Mac for it here is an M5 Ultra with 256 GB at about 74–132 tokens a second (4-bit). The smallest Mac that runs it, with 256 GB, manages about 74–132 on an M5 Ultra.
- Why does DeepSeek V4 Flash 0731 run faster than its size suggests?
- It is a mixture-of-experts model: it holds 284B parameters in memory but reads only 13B for each token, so it writes at the speed of a much smaller model.
Similar models
For your model card or README
Shows the smallest Mac that runs DeepSeek V4 Flash 0731 well and links here.
[](https://mlx.app/run/deepseek-v4-flash-0731)How it’s worked out
Memory: the model’s parameters at 8, 6, 4, 3 or 2 bits per weight plus MLX’s group scales (half a bit per weight), or its native format, with room for the context on top. By default macOS lets the GPU use about 75% of unified memory. Speed: memory bandwidth divided by the bytes read per token is the ceiling for one conversation; in llama.cpp’s public Apple Silicon benchmarks real runs reach about 45% to 80% of it, which is the range shown. These are estimates, not measurements: measure your own Mac on MLX Bench. Model figures from each official model card, chip figures from Apple’s Tech Specs. Checked 2026-10-19. Same data as mlx.app/compare.