Feature Extraction
Transformers
Safetensors
fast_esm3
biology
protein-language-model
esm3
multimodal-protein-model
custom_code
Instructions to use Synthyra/ESM3_small with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Synthyra/ESM3_small with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="Synthyra/ESM3_small", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Synthyra/ESM3_small", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "FastESM3Model" | |
| ], | |
| "attn_backend": "sdpa", | |
| "auto_map": { | |
| "AutoConfig": "modeling_esm3.FastESM3Config", | |
| "AutoModel": "modeling_esm3.FastESM3Model", | |
| "AutoModelForMaskedLM": "modeling_esm3.FastESM3Model" | |
| }, | |
| "hidden_size": 1536, | |
| "initializer_range": 0.02, | |
| "model_name": "esm3_sm_open_v1", | |
| "model_type": "fast_esm3", | |
| "num_attention_heads": 24, | |
| "num_hidden_layers": 48, | |
| "num_vector_heads": 256, | |
| "tie_word_embeddings": false, | |
| "transformers_version": "4.57.6", | |
| "vocab_size": 64 | |
| } | |