Falcon is back! Similar to Nemotron-H, they use a hybrid architecture consisting of Attention and Mamba2 blocks for faster inference while needing less memory.
Specs
Params0.5B, 1.5B, 3B, 7B, 34B
LicenseFalcon-LLM License
Tags
modelsmodels/llmsmultilingualmodels/llms/reasoningartifacts/10
RAM score
—
Hugging Face Downloads
1K
last 30d
48.5K
all time
HF Likes
52
Relative Adoption Metric contextualizes downloads against the model's size bucket.
VAIL Fingerprint
0053:0066:00bd:0108:0191:02da:06bd:58e1
Explore other models with behavioral similarity to Falcon-H1-34B-Instruct.
Related Models
