Memory at each precision
| Version | Weights | With context | Smallest Mac |
|---|---|---|---|
| 8-bit | 191 GB | 207 GB | 512 GB |
| 6-bit | 146 GB | 158 GB | 256 GB |
| 4-bit | 101 GB | 109 GB | 192 GB |
| 3-bit lower quality | 78.8 GB | 85.1 GB | 128 GB |
| 2-bit lower quality | 56.3 GB | 60.8 GB | 96 GB |
On every Apple chip
The smallest memory size of each chip that runs Qwen3.8 Flash Next, 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 | 6-bit Runs well | 110–196 |
| M5 Max | 128 GB | 3-bit Runs well | 106–189 |
| M5 Pro | no size fits | Too big | – |
| M5 | no size fits | Too big | – |
| M4 Max | 128 GB | 3-bit Runs well | 95–168 |
| M4 Pro | no size fits | Too big | – |
| M4 | no size fits | Too big | – |
| M3 Ultra | 256 GB | 6-bit Runs well | 75–134 |
| M3 Max | 96 GB | 2-bit Runs well | 71–126 |
| M3 Pro | no size fits | Too big | – |
| M3 | no size fits | Too big | – |
| M2 Ultra | 192 GB | 4-bit Runs well | 106–188 |
| M2 Max | 96 GB | 2-bit Runs well | 95–168 |
| M2 Pro | no size fits | Too big | – |
| M2 | no size fits | Too big | – |
| M1 Ultra | 128 GB | 3-bit Runs well | 138–246 |
| 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 Qwen3.8 Flash Next, 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/Qwen3.8-Flash-Next-4bitBuild: mlx-community/Qwen3.8-Flash-Next-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 Qwen3.8 Flash Next
| Made by | Qwen (Alibaba) |
|---|---|
| Parameters | 180B total · 6B active (mixture of experts) |
| Context | 256K tokens |
| Licence | Qwen Community License 1.0 (qwen-community-1.0) |
| Released | 2026-08 |
| Input | Text and images |
| Intelligence | 39.8 on the Artificial Analysis Intelligence Index |
| Weights | Qwen/Qwen3.8-Flash-Next |
Questions
- How much memory does Qwen3.8 Flash Next need?
- About 101 GB of weights at 4-bit, plus room for the context: 109 GB in all. On a Mac that means 192 GB of unified memory or more, since macOS lets the GPU use about 75% of it by default.
- How fast is Qwen3.8 Flash Next on a Mac?
- Writing speed depends on memory bandwidth. The fastest Mac for it here is an M5 Ultra with 256 GB at about 110–196 tokens a second (6-bit). The smallest Mac that runs it, with 192 GB, manages about 106–188 on an M2 Ultra.
- Why does Qwen3.8 Flash Next run faster than its size suggests?
- It is a mixture-of-experts model: it holds 180B parameters in memory but reads only 6B 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 Qwen3.8 Flash Next well and links here.
[](https://mlx.app/run/qwen3.8-flash-next)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.