stepfun-ai · First MLX build 2 Feb 2026
Step-3.5-Flash
MLX
Text apache-2.0
A 197B parameter text generation model from stepfun-ai, with 4 MLX builds, from 4-bit to bf16.
- Parameters
- 197B From Hugging Face
- Quantisation
- 4-bit Step-3.5-Flash-4bit
- Memory needed
- 126 GB Estimate, unified memory
- Download
- 111 GB Weights, estimated
- Downloads
- 423 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 4 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 126 GB at 4-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Step-3.5-Flash-4bit --prompt "Hello"MLX builds
4 on mlx-community| Step-3.5-Flash-4bit | 4-bit | 111 GB | 126 GB | 423 |
|---|---|---|---|---|
| Step-3.5-Flash-6bit | 6-bit | 160 GB | 181 GB | 87 |
| Step-3.5-Flash-8bit | 8-bit | 209 GB | 236 GB | 91 |
| Step-3.5-Flash-bf16 | bf16 | 394 GB | 443 GB | 27 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.