tiiuae / text-generation updated 1 year ago

falcon-mamba-7b

Name: falcon-mamba-7b
Author: tiiuae

- Developed by: https://www.tii.ae - Model type: Causal decoder-only - Architecture: Mamba - Language(s) (NLP): Mainly English - License: TII Falcon-Mamba License 2.0

Params

7.3 B

Context

—

Downloads 30d

174 K

Likes

244

Licence: other Not gated SAFETENSORS 1 languages View on Hugging Face ↗

Download history

daily snapshots · 10 days

178 K165 K

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	bf16	14.5 GB	17.6 GB	✅ Runs comfortably

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub

~ · curl · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/falcon-mamba-7b
{
  "hf_id": "tiiuae/falcon-mamba-7b",
  "params_b": 7.27,
  "context_length": null,
  "license": { "id": "other", "commercial": "unknown" },
  "downloads_30d": 173837,
  "vram_estimates": [
    { "quant": "bf16", "gb": 17.6 }
  ],
  "updated_at": "2026-07-28T18:08:34Z"
}

est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture: FalconMambaForCausalLM
Parameters: 7.3 B
Tensor type: BF16
Vocabulary: 65,024
Licence: other
First seen on the Hub: 2024-07-17
Training datasets: tiiuae/falcon-refinedweb, HuggingFaceFW/fineweb-edu
BBH (3-Shot) (reported): 19.88
GPQA (0-shot) (reported): 8.05
MuSR (0-shot) (reported): 10.86
IFEval (0-Shot) (reported): 33.36
MMLU-PRO (5-shot) (reported): 14.47
MATH Lvl 5 (4-Shot) (reported): 3.63
Added to our catalog: 2026-07-28

Compare with

Qwen3-8B 16.3 M Qwen2.5-7B-Instruct 12.1 M Qwen3.6-35B-A3B-NVFP4 11.3 M Llama-3.1-8B-Instruct 8.0 M Qwen2.5-7B-Instruct-AWQ 4.6 M granite-4.1-8b 4.2 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB