Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
kdirgul
/
Mamba3-177M-GQA-Hybrid_LAMBA_V1.0
like
3
Text Generation
English
Turkish
mamba_ssm
mamba
mamba-3
state-space-model
gqa
hybrid
turkish
bilingual
rag
causal-lm
License:
lamba-v1-license-tbd
Model card
Files
Files and versions
xet
Community
Copy to bucket
new
main
Mamba3-177M-GQA-Hybrid_LAMBA_V1.0
811 MB
Ctrl+K
Ctrl+K
1 contributor
History:
14 commits
kdirgul
Update README.md
193ef31
verified
about 1 month ago
checkpoints
LAMBA V1.0 weights
about 1 month ago
tokenizer
Upload tokenizer/tokenizer.model with huggingface_hub
about 1 month ago
wheels
Mamba-3 GPU wheels
about 1 month ago
.gitattributes
Safe
1.71 kB
Mamba-3 GPU wheels
about 1 month ago
LAMBA_Inference.ipynb
Safe
4.11 kB
v1.1: CPU support — pure-PyTorch lamba_cpu.py (decode-cache, no Triton) + lamba_rag.py --device cpu + README/notebook
about 1 month ago
README.md
Safe
7.74 kB
Update README.md
about 1 month ago
lamba_cpu.py
Safe
19.5 kB
v1.1.x: optional --int8 dynamic quant (CPU, ~2x smaller memory 708->325MB, identical output) + README note
about 1 month ago
lamba_rag.py
Safe
18.7 kB
v1.1.x: optional --int8 dynamic quant (CPU, ~2x smaller memory 708->325MB, identical output) + README note
about 1 month ago
