What are the hardware requirements for Falcon Mamba 7B?

Falcon Mamba 7B requires approximately 14.6 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Falcon Mamba 7B and how large is it?

Falcon Mamba 7B is developed by tiiuae, a model with 7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Falcon Mamba 7B?

Falcon Mamba 7B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Falcon Mamba 7B by tiiuae — VRAM 14.6GB

Name: Falcon Mamba 7B
Author: tiiuae

Falcon Mamba 7B is an open-source language model by tiiuae. Features: 7b LLM, VRAM: 14.6GB, License: other, LLM Explorer Score: 0.26.

Arxiv:2312.00752 Arxiv:2410.05355 Dataset:huggingfacefw/fineweb-... Dataset:tiiuae/falcon-refinedw... En Endpoints compatible Falcon mamba Model-index Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/tiiuae/falcon-mamba-7b

Falcon Mamba 7B Parameters and Internals

Model Type

Causal decoder-only

Use Cases

Areas:

intended for text generation

Supported Languages

English (Main language)

Training Details

Data Sources:

tiiuae/falcon-refinedweb, HuggingFaceFW/fineweb-edu

Data Volume:

~ 5,500 GT

Methodology:

multi-stage training to increase context-length, Curriculum Learning

Context Length:

8192

Training Time:

roughly two months

Hardware Used:

256 H100 80GB GPUs

Model Architecture:

Mamba architecture

Input Output

Performance Tips:

Install optimized Mamba kernels with 'causal-conv1d>=1.4.0' and 'mamba-ssm'.

LLM Name	Falcon Mamba 7B
Repository 🤗	https://huggingface.co/tiiuae/falcon-mamba-7b
Model Size	7b
Required VRAM	14.6 GB
Updated	2026-06-24
Maintainer	tiiuae
Model Type	falcon_mamba
Model Files	5.0 GB: 1-of-3 5.0 GB: 2-of-3 4.6 GB: 3-of-3
Supported Languages	en
Model Architecture	FalconMambaForCausalLM
License	other
Transformers Version	4.43.0.dev0
Tokenizer Class	PreTrainedTokenizerFast
Vocabulary Size	65024
Torch Data Type	bfloat16

Rank the Falcon Mamba 7B capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Quantized Models of the Falcon Mamba 7B

Model	Likes	Downloads	VRAM
Falcon Mamba 7B 4bit	11	16	4 GB

Best Alternatives to Falcon Mamba 7B

Best Alternatives	Context / RAM	Downloads	Likes
Falcon3 Mamba 7B Instruct	0K / 14.6 GB	936	33
Falcon3 Mamba 7B Base	0K / 14.6 GB	234	24
Falcon Mamba 7B Instruct	0K / 14.6 GB	18925	73
Falcon3 Mamba R1 V0	0K / 14.6 GB	11	12
Falcon Mamba 7B 4bit	0K / 4.9 GB	16	11