Audio-Text-to-Text
Transformers
Safetensors
Russian
English
borealis
feature-extraction
audio
speech
multimodal
whisper
qwen
custom_code
Instructions to use Vikhrmodels/Borealis-5b-it with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Vikhrmodels/Borealis-5b-it with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Vikhrmodels/Borealis-5b-it", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": ["BorealisForConditionalGeneration"], | |
| "model_type": "borealis", | |
| "whisper_model_name": "openai/whisper-large-v3", | |
| "llm_model_name": "Qwen/Qwen3-4B", | |
| "downsample_factor": 4, | |
| "audio_hidden_size": 1280, | |
| "llm_hidden_size": 2560, | |
| "hidden_size": 2560, | |
| "vocab_size": 151671, | |
| "torch_dtype": "bfloat16", | |
| "text_config": { | |
| "model_type": "qwen3", | |
| "hidden_size": 2560, | |
| "intermediate_size": 6912, | |
| "num_hidden_layers": 36, | |
| "num_attention_heads": 32, | |
| "num_key_value_heads": 4, | |
| "max_position_embeddings": 40960, | |
| "rope_theta": 1000000.0, | |
| "rms_norm_eps": 1e-6, | |
| "vocab_size": 151671, | |
| "bos_token_id": 151643, | |
| "eos_token_id": 151645, | |
| "head_dim": 80, | |
| "tie_word_embeddings": true | |
| }, | |
| "auto_map": { | |
| "AutoConfig": "modeling_borealis.BorealisConfig", | |
| "AutoModel": "modeling_borealis.BorealisForConditionalGeneration", | |
| "AutoModelForCausalLM": "modeling_borealis.BorealisForConditionalGeneration", | |
| "AutoProcessor": "processing_borealis.BorealisProcessor" | |
| }, | |
| "transformers_version": "4.48.0" | |
| } | |