See DeepSeek-V3.2 MLX in action - demonstration video
q4.8bit mixed quant typically achieves 1.14 perplexity in our testing
| Quantization | Perplexity |
|---|---|
| q2.5 | 41.293 |
| q3.5 | 1.900 |
| q4.5 | 1.168 |
| q4.8 | 1.140 |
| q5.5 | 1.141 |
| q6.5 | 1.128 |
| q8.5 | 1.128 |
Usage Notes
Tested by remotely connecting to a M3 Ultra 512GB RAM using Inferencer app v1.7.3
- Expect ~18 tokens/s @ 1000 tokens
- Memory usage: ~382 GB
Quantized with a modified version of MLX 0.28
For more details see demonstration video or visit DeepSeek-V3.2.
Disclaimer
We are not the creator, originator, or owner of any model listed. Each model is created and provided by third parties. Models may not always be accurate or contextually appropriate. You are responsible for verifying the information before making important decisions. We are not liable for any damages, losses, or issues arising from its use, including data loss or inaccuracies in AI-generated content.
- Downloads last month
- 1,107
Model size
672B params
Tensor type
BF16
·
U32 ·
F32 ·
Hardware compatibility
Log In to add your hardware
4-bit
Model tree for inferencerlabs/DeepSeek-V3.2-MLX-4.8bit
Base model
deepseek-ai/DeepSeek-V3.2-Exp-Base Finetuned
deepseek-ai/DeepSeek-V3.2