Goekdeniz-Guelmez · First MLX build 17 Nov 2024
Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1
MLX
Text apache-2.0
A 494M parameter text generation model from Goekdeniz-Guelmez, with 5 MLX builds, from 4-bit to 8-bit.
- Parameters
- 494M From Hugging Face
- Quantisation
- Unstated Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1-float32
- Memory needed
- 5 GB Estimate, unified memory
- Download
- 2.0 GB Weights, estimated
- Downloads
- 114 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 5 GB at Unstated
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/Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1-float32 --prompt "Hello"MLX builds
5 on mlx-community| Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1-4bit | 4-bit | 278 MB | 3 GB | 61 |
|---|---|---|---|---|
| Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1-8bit | 8-bit | 525 MB | 3 GB | 52 |
| Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1 | Unstated | 988 MB | 4 GB | 105 |
| Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1-float16 | Unstated | 988 MB | 4 GB | 43 |
| Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1-float32 | Unstated | 2.0 GB | 5 GB | 114 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.