Instructions to use havenhq/mamba-chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use havenhq/mamba-chat with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("havenhq/mamba-chat", device_map="auto") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from havenhq/mamba-chat: direct link, hf CLI and curl.
- Browser
- Download file 615 Bytes
-
https://huggingface.co/havenhq/mamba-chat/resolve/refs%2Fpr%2F1/README.md
- Command line
-
hf download hf://havenhq/mamba-chat@refs/pr/1/README.md
-
curl -L -o README.md https://huggingface.co/havenhq/mamba-chat/resolve/refs%2Fpr%2F1/README.md
615 Bytes
metadata
license: apache-2.0
Mamba-Chat is the first chat language model based on a state-space model architecture, not a transformer.
The model is a fine-tune of Albert Gu's and Tri Dao's model Mamba-2.8B from their paper Mamba: Linear-Time Sequence Modeling with Selective State Spaces.
Check our our Github repository for training and inference code.
The prompt format is the zephyr format:
<|user|> {user_message}
<|assistant|> {assistant_message}
<|user|> {user_message}
<|assistant|> {assistant_message}