Downloads · 30 days
37
13% of all-time downloads
Qwen3.8-27B-oQ4e-mtp
jwheaton/Qwen3.8-27B-oQ4e-mtp
Qwen3.8-27B-oQ4e-mtp is a machine learning model from jwheaton. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for mlx.
This model was quantized using oQ (oMLX v0.6.0.dev1) mixed-precision quantization.
All-time downloads
276
Public
Parameters
27.8B
17 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors17 GB · 100%
Parameter types
How the weights are stored.
U3227.3B · 98%
Model card
From the Hugging Face model README
Qwen--Qwen3.8-27B-oQ4e-mtp
This model was quantized using oQ (oMLX v0.6.0.dev1) mixed-precision quantization.
Quantization details
- Model type: qwen3_5
- Bits: 4
- Group size: 64
- Format: MLX safetensors
Intelligence Benchmark Comparison
| Benchmark | Mode | Sampled | Qwen--Qwen3.8-27B-oQ4e-mtp |
|---|---|---|---|
| MMLU | Sample | 1000/14042 | 82.9% |
| WINOGRANDE | Sample | 300/1267 | 78.0% |
| HUMANEVAL | Full | 164 | 91.5% |
| MBPP | Sample | 300/500 | 81.0% |
Benchmark Details
Model: Qwen--Qwen3.8-27B-oQ4e-mtp
| Benchmark | Accuracy | Correct | Total | Time (s) | Think |
|---|---|---|---|---|---|
| MMLU | 82.9% | 829 | 1000 | 1892.6 | No |
| WINOGRANDE | 78.0% | 234 | 300 | 134.3 | No |
| HUMANEVAL | 91.5% | 150 | 164 | 600.6 | No |
| MBPP | 81.0% | 243 | 300 | 970.0 | No |
- Library
- mlx
- Type
- qwen3_5
- Created
- Aug 14, 2026
- Updated
- Aug 14, 2026