Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-4-bit
mlx-community/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-4-bit
Text generation · mlx
4-bit Open on Hugging Face
Parameters
7.6B
Weights (est.)
≈ 4.3 GB
Mac memory (est.)
8 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-4-bit --prompt "Hello"
Details
- Base model
- Goekdeniz-Guelmez/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2 (quantized)
- Vendor
- Goekdeniz-Guelmez
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 315 in the last 30 days · 5.5K all time · 3 likes
- Published
- 2024-09-29 · updated 2024-12-08 (1y ago)
- Languages
- en
- Tags
- qwen2, chat, model-index
Other MLX builds of Goekdeniz-Guelmez/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-2-bit | 2-bit | ≈ 2.4 GB | 8 GB+ | 94 |
| Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-4-bit (this one) | 4-bit | ≈ 4.3 GB | 8 GB+ | 315 |
| Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-8-bit | 8-bit | ≈ 8.1 GB | 16 GB+ | 53 |
| Josiefied-Qwen2.5-7B-Instruct-abliterated-v2 | — | ≈ 15.2 GB | 24 GB+ | 201 |