EXL3 Quants of Trappu/Magnum-Picaro-0.7-v2-12b
EXL3 quants of Trappu/Magnum-Picaro-0.7-v2-12b using exllamav3 for quantization.
Quants
| Quant(Revision) | Bits per Weight | Head Bits |
|---|---|---|
| 2.5_H6 | 2.5 | 6 |
| 3.0_H6 | 3.0 | 6 |
| 3.5_H6 | 3.5 | 6 |
| 4.0_H6 | 4.0 | 6 |
| 4.25_H6 | 4.25 | 6 |
| 4.5_H6 | 4.5 | 6 |
| 5.0_H6 | 5.0 | 6 |
| 5.5_H6 | 5.5 | 6 |
| 6.0_H6 | 6.0 | 6 |
| 8.0_H8 | 8.0 | 8 |
Downloading quants with huggingface-cli
Click to view download instructions
Install hugginface-cli:
pip install -U "huggingface_hub[cli]"
Download quant by targeting the specific quant revision (branch):
huggingface-cli download ArtusDev/Trappu_Magnum-Picaro-0.7-v2-12b-EXL3 --revision "5.0bpw_H6" --local-dir ./
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for ArtusDev/Trappu_Magnum-Picaro-0.7-v2-12b-EXL3
Base model
Trappu/Magnum-Picaro-0.7-v2-12bEvaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard30.030
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard35.750
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard4.760
- acc_norm on GPQA (0-shot)Open LLM Leaderboard9.730
- acc_norm on MuSR (0-shot)Open LLM Leaderboard19.560
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard28.670