Instructions to use yxdu/ESRT-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use yxdu/ESRT-4B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="yxdu/ESRT-4B", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("yxdu/ESRT-4B", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "CustomSLM" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "srt_model.CustomSLMConfig", | |
| "AutoModel": "srt_model.CustomSLM" | |
| }, | |
| "dtype": "bfloat16", | |
| "encoder_dim": 1280, | |
| "encoder_only": false, | |
| "encoder_path": "yxdu/whisper-large-v3-encoder", | |
| "llm_dim": 2560, | |
| "llm_path": "yxdu/ESRT-4B-LLM", | |
| "model_type": "custom_slm", | |
| "qformer_layers": 8, | |
| "query_len": 80, | |
| "tie_word_embeddings": false, | |
| "tokenizer_path": "yxdu/ESRT-4B-LLM", | |
| "transformers_version": "4.57.3", | |
| "use_vllm": false, | |
| "vllm_path": "yxdu/ESRT-4B-LLM" | |
| } | |