jdopensource · First MLX build 15 Feb 2026
JoyAI-LLM-Flash
MLX
Text
A 49B parameter text generation model from jdopensource, with 5 MLX builds, from 4-bit DWQ to 8-bit.
- Parameters
- 49B From Hugging Face
- Quantisation
- 4-bit DWQ JoyAI-LLM-Flash-4bit-DWQ
- Memory needed
- 33 GB Estimate, unified memory
- Download
- 27.6 GB Weights, estimated
- Downloads
- 43 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 33 GB at 4-bit DWQ
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/JoyAI-LLM-Flash-4bit-DWQ --prompt "Hello"MLX builds
5 on mlx-community| JoyAI-LLM-Flash-4bit-DWQ | 4-bit DWQ | 27.6 GB | 33 GB | 43 |
|---|---|---|---|---|
| JoyAI-LLM-Flash-4bit | 4-bit | 27.6 GB | 33 GB | 26 |
| JoyAI-LLM-Flash-5bit | 5-bit | 33.7 GB | 40 GB | 23 |
| JoyAI-LLM-Flash-6bit | 6-bit | 39.8 GB | 47 GB | 29 |
| JoyAI-LLM-Flash-8bit | 8-bit | 52.0 GB | 61 GB | 27 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.