Moomboh/ESMC-300M-mutafitup
By Moomboh · ↓ 0 · ♥ 0 · Updated 2026-03-14
What's in this repo
This page lists the files in Moomboh/ESMC-300M-mutafitup, including model.onnx and more.
- 2 ONNX models (.onnx files)
- 834 files, including 418 .pt, 105 .json, 2 .md, 2 .onnx
- Last updated 2026-03-14
ONNX files
| File | Size | Download |
|---|---|---|
| ESMC-300M-mutafitup-accgrad-all-r4-best-overall/model.onnx | — | Download |
| ESMC-300M-mutafitup-align-all-r4-best-overall/model.onnx | — | Download |
How to use
These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.
README
Moomboh/ESMC-300M-mutafitup
Multi-task LoRA fine-tuned ONNX models derived from ESM-C 300M by EvolutionaryScale.
Built with ESM.
ONNX Models
| Model | Section | Tasks | Variant |
|---|---|---|---|
ESMC-300M-mutafitup-accgrad-all-r4-best-overall |
accgrad_lora | disorder, gpsite_atp, gpsite_ca, ... (16 total) | best_overall |
ESMC-300M-mutafitup-align-all-r4-best-overall |
align_lora | disorder, gpsite_atp, gpsite_ca, ... (16 total) | best_overall |
Each ONNX model directory contains:
model.onnx-- merged ONNX model (LoRA weights folded into backbone)export_metadata.json-- task configuration and preprocessing settingsnormalization_stats.json-- per-task normalization statisticstokenizer/-- HuggingFace tokenizer fileshistory.json-- training history (per-epoch metrics)best_checkpoints.json-- checkpoint selection metadata
PyTorch Checkpoints
The checkpoints/ directory contains minimal trainable-parameter
PyTorch checkpoints for all training runs (45 runs across
4 training sections). These checkpoints contain only the
parameters that were updated during fine-tuning (LoRA adapters and task
heads), not the frozen backbone weights.
Each run directory (checkpoints/{section}/{run}/) contains:
history.json-- training historybest_checkpoints.json-- checkpoint selection metadatabest_overall_model/model.pt-- best checkpoint by overall metricbest_loss_overall_model/model.pt-- best checkpoint by overall lossbest_task_models/{task}/model.pt-- best checkpoint per task metricbest_loss_task_models/{task}/model.pt-- best checkpoint per task loss
To load a checkpoint, use MultitaskModel.load_trainable_weights() from
the mutafitup training library.
License
The ESMC 300M base model is licensed under the EvolutionaryScale Cambrian Open License Agreement.
Fine-tuning code and pipeline are licensed under the MIT License.
See NOTICE for full attribution details.
Model info
Created: 2026-03-14
Updated: 2026-03-14
View on Hugging FaceFiles (834)
Related models
More from Moomboh
No other models by this author.
More models
No task type declared.