EXL3 quantization of Nanbeige4-3B-Thinking-2511, 8 bits per weight, including output layers.
HumanEval (argmax)
| Model | Q4 | Q6 | Q8 | FP16 |
|---|---|---|---|---|
| Nanbeige4-3B-Thinking-2511-exl3-8bpw-h8 | 65.2 | 70.1 | 70.1 | 68.9 |
- Downloads last month
- 2
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for isogen/Nanbeige4-3B-Thinking-2511-exl3-8bpw-h8
Base model
Nanbeige/Nanbeige4-3B-Base Finetuned
Nanbeige/Nanbeige4-3B-Thinking-2511