Supermix V33 Frontier

v33_final is the best saved common-benchmark text model from the Supermix v33 frontier line in this workspace.

Files

  • champion_model_chat_v33_frontier_full_final.pth
  • chat_model_meta_v33_frontier_full_final.json
  • v33_frontier_training_summary.json
  • model_frontier_v33.py
  • model_variants.py
  • run.py

Important Note

This is a custom PyTorch checkpoint, not a standard Transformers from_pretrained model.

Local Sources

  • weights and metadata were extracted from champion_v33_frontier_full_model_20260326.zip
  • architecture code comes from the local Supermix source tree

Benchmark Context

Saved expanded common-benchmark score in this workspace:

  • overall exact: 0.1867
  • ARC-Challenge: 0.20
  • BoolQ: 0.08
  • GSM8K: 0.02
  • HellaSwag: 0.26
  • MMLU: 0.14
  • PIQA: 0.42

Usage

This checkpoint needs the included Python source files to construct the model class before loading the .pth weights.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support