Audio-Text-to-Text
Transformers
Safetensors
English
Korean
fastslm
feature-extraction
audio
text-generation
custom_code
Eval Results
Instructions to use okestro-ai-lab/FastSLM-ASR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use okestro-ai-lab/FastSLM-ASR with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("okestro-ai-lab/FastSLM-ASR", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "FastSLMForConditionalGeneration" | |
| ], | |
| "encoder_config": { | |
| "compression_size": 50, | |
| "model_type": "fastslm_speech_encoder", | |
| "n_ctx": 1500, | |
| "n_head": 20, | |
| "n_layer": 32, | |
| "n_mels": 128, | |
| "n_state": 1280, | |
| "stage_tokens": [ | |
| 80, | |
| 80, | |
| 80 | |
| ] | |
| }, | |
| "llm_config": { | |
| "_name_or_path": "Qwen/Qwen3-4B", | |
| "architectures": [ | |
| "Qwen3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 151643, | |
| "eos_token_id": 151645, | |
| "head_dim": 128, | |
| "hidden_act": "silu", | |
| "hidden_size": 2560, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 9728, | |
| "layer_types": [ | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention" | |
| ], | |
| "max_position_embeddings": 40960, | |
| "max_window_layers": 36, | |
| "model_type": "qwen3", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 36, | |
| "num_key_value_heads": 8, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_parameters": { | |
| "rope_type": "default", | |
| "rope_theta": 1000000 | |
| }, | |
| "sliding_window": null, | |
| "tie_word_embeddings": true, | |
| "torch_dtype": "bfloat16", | |
| "use_cache": true, | |
| "use_sliding_window": false, | |
| "vocab_size": 151936 | |
| }, | |
| "llm_modules": [ | |
| "q_proj", | |
| "k_proj", | |
| "v_proj", | |
| "o_proj", | |
| "gate_proj", | |
| "up_proj", | |
| "down_proj" | |
| ], | |
| "lora_a": 64, | |
| "lora_r": 16, | |
| "low_resource": false, | |
| "model_type": "fastslm", | |
| "auto_map": { | |
| "AutoConfig": "configuration_fastslm.FastSLMConfig", | |
| "AutoModel": "modeling_fastslm.FastSLMForConditionalGeneration", | |
| "AutoModelForCausalLM": "modeling_fastslm.FastSLMForConditionalGeneration" | |
| }, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.53.1" | |
| } |