We value your privacy

We use cookies to enhance your browsing experience, serve personalized content and analyze our traffic. By clicking “Accept all” you consent to our use of cookies. Cookie Policy

More options
Skip to content
FacthoryFacthory
FacthoryFacthory
    • Workspace
    • Agent
    • Knowledge
    • Analytics
    • Office apps
    • Desktop
    All features
    What’s new
    Download
  • By work

    • Organizational Knowledge Risk & Gap Intelligence
    • Scientific Research & Experimentation
    • Autonomous Analytics & Insight Discovery
    • Workforce Readiness & Adaptive Learning
    • SOP Governance & Operational Standardization
    • Process Discovery & Continuous Optimization
    • All solutions

    By industry

    • Manufacturing
    • Automotive
    • Construction & Infrastructure
    • Real Estate
    • Public Sector
    • Life Sciences
    • All industries
  • Facthory for

    • Regulated industries
    • On-premises
    • Developers

    Discover

    • What’s new
    • Blog
    • Case studies
    • Webinars
    • Models

    Learn

    • Documentation
    • Help centre
  • Pricing
  • Download
Sign inGet started

A workspace for what your organization knows, with an agent that works on it.

Product

  • All features
  • Workspace
  • Agent
  • Knowledge
  • Analytics
  • Office apps
  • Desktop
  • Download
  • Pricing

Solutions

  • Organizational Knowledge Risk & Gap Intelligence
  • Scientific Research & Experimentation
  • Autonomous Analytics & Insight Discovery
  • Workforce Readiness & Adaptive Learning
  • SOP Governance & Operational Standardization
  • Process Discovery & Continuous Optimization
  • All solutions
  • All industries

Resources

  • Blog
  • Case studies
  • Webinars
  • Documentation
  • Help centre
  • Models
  • ROI calculator

Company

  • Who we are
  • Careers
  • Press hub
  • Brand kit
  • Philosophy
  • Contact us

Join our newsletter

Product news from Facthory, by email.

​

© 2026 Facthory. All rights reserved.
  • Privacy policy
  • Cookie policy
  • Terms and conditions
  • Imprint
  • Cookie settings
All modelstiiuae

falcon-mamba-7b

text-generation
7.3B parameters
other
No file flagged

Falcon-Mamba-7B is a causal decoder-only model developed by TII, trained mainly on English data from Refined-Web and FineWeb-Edu to handle long context lengths.

View on Hugging FaceCompare with other models

Fact sheet

Parameters

7.3B

Context length

not published

Architecture

FalconMambaForCausalLM

Shape

not published

Weights stored as

bfloat16

Task

text-generation

Languages

en

Built on

–

Released

17 Jul 2024

Last updated

17 Dec 2024

Downloads (last 30 days)

4,900

Licence and access

Licence

other (falcon-mamba-7b-license) · Read the licence

Access

Weights download without approval

Security

Hugging Face security scan

No file flagged

Weight files by format

Safetensors: 3
Pickle-based: 0
GGUF: 0
Other formats: 0
Pickle-based files (.bin, .pt, .pth, .pkl, .ckpt) can run code when they are loaded; safetensors files cannot.

Facthory’s checks

Automated
Met
Safe weight formats

Every weight file is safetensors or GGUF.

Met
The hub’s malware and pickle scan

No file flagged.

Met
A declared licence

The card declares its licence.

Not evidenced
Published safety evaluations

GPU calculator

How much memory this model needs to serve at a quantization, context and number of concurrent sequences, which GPUs hold it, and how fast it can generate. Estimates from the model’s published size and shape, with the assumptions below.

Weights

BF16 / FP16 (16 bit)
​

Context per sequence: 8K tokens

Concurrent sequences: 1

KV cache

Memory

16 GB

Weights: 15 GB

KV cache: –

Runtime: 1.5 GB

The model’s shape is not published, so the KV cache is not included.

GPUMemoryGPUs neededPer sequenceAll sequences
NVIDIA L424 GB
1 ×
12 tokens/s12 tokens/s
NVIDIA L40S48 GB
1 ×
36 tokens/s36 tokens/s
NVIDIA RTX 6000 Ada48 GB
1 ×
40 tokens/s40 tokens/s
NVIDIA RTX PRO 6000 Blackwell96 GB
1 ×
74 tokens/s74 tokens/s
NVIDIA A100 80GB SXM80 GB
1 ×
84 tokens/s84 tokens/s
NVIDIA H100 SXM80 GB
1 ×
138 tokens/s138 tokens/s
NVIDIA H200 SXM141 GB
1 ×
198 tokens/s198 tokens/s
AMD Instinct MI300X192 GB
1 ×
219 tokens/s219 tokens/s
AMD Instinct MI325X256 GB
1 ×
248 tokens/s248 tokens/s
Assumptions
  • Runtime memory (activations, buffers) is 10% of weights and KV cache, plus 1 GB per GPU for the framework.
  • A GPU is filled to at most 90% of its memory; tensor parallelism uses 1, 2, 4 or 8 GPUs in one server.
  • Generation is limited by memory bandwidth at 60% of the vendor’s peak: each step reads the weights once and every sequence’s KV cache. A mixture-of-experts model reads only its active experts, so it runs faster than shown.
  • GPU memory and bandwidth are the vendors’ published figures.

Profile

Kind

Instruction model

Lineage

Original

Family

FalconMamba

Uses

Coding
Math
Reasoning
Science
Conversation

In and out

Text → Text

Written on 12 Nov 2026 by Workers AI (@cf/zai-org/glm-4.7-flash) from the model card and the hub’s metadata: a reading of the publisher’s text, not checked by Facthory.

Published benchmarks

As the model card publishes them, with the source the card names. Facthory has not run them; results are self-reported unless the hub verified them.

BenchmarkMetricResultSourceCounts toward
BBH (3-Shot)
Not counted: the leaderboard’s run
normalized accuracy19.88Open LLM LeaderboardNot counted: the leaderboard’s run
GPQA (0-shot)
Not counted: the leaderboard’s run
acc_norm8.05Open LLM LeaderboardNot counted: the leaderboard’s run
IFEval (0-Shot)
Not counted: the leaderboard’s run
strict accuracy33.36Open LLM LeaderboardNot counted: the leaderboard’s run
MATH Lvl 5 (4-Shot)
Not counted: the leaderboard’s run
exact match3.63Open LLM Leaderboard

EU AI Act

What the EU AI Act asks of a general-purpose AI model depends on who places it on the EU market and how. These facts bear on it:

  • Licence: other
  • Access: Weights download without approval
  • Parameters: 7.3B
Automated
Rated by Facthory’s crawler from the published facts; a person has not reviewed it yet.

General-purpose AI model (Article 3(63))

Not determined. No estimate of its training compute is published, and the Commission’s indicative criterion (10²³ FLOP) needs one. If it is a general-purpose AI model, the obligations below bind its provider.

Systemic risk (Article 51)

Not presumed. No estimate of its training compute is published; without one, or a designation by the Commission under Article 52, Article 55 does not apply.

Open-source exemption (Article 53(2))

Does not apply. Its licence (other) is not a free and open-source licence, one that lets anyone use, study, change and share the model.

General-Purpose AI Code of Practice

Technology Innovation Institute is not on the European Commission’s list of signatories. The Code is voluntary.

What the card and the provider document

Met
Technical documentation published (Article 53(1)(a), Annex XI)

The card publishes the technical documentation’s items.

Not stated: energy used.
Not evidenced
Information for downstream providers (Article 53(1)(b), Annex XII)

The card gives almost none of the items downstream providers need.

Source: Read from Hugging Face on 19 Nov 2026, at commit 080ad94. huggingface.co/tiiuae/falcon-mamba-7b

The card reports no safety evaluation.

Rated by Facthory’s crawler from the published facts; a person has not reviewed it yet. How we rate
Not counted: the leaderboard’s run
MMLU-PRO (5-shot)
main · test
Not counted: the leaderboard’s run
accuracy14.47Open LLM LeaderboardNot counted: the leaderboard’s run
MuSR (0-shot)
Not counted: the leaderboard’s run
acc_norm10.86Open LLM LeaderboardNot counted: the leaderboard’s run

The capability score counts results the model’s authors publish, for the benchmarks our methodology lists, where at least two models in the catalog report the same benchmark, configuration and metric. Results a card attributes to the Open LLM Leaderboard are the leaderboard’s runs, and Epoch AI’s evaluations are Epoch’s: neither counts. How we score capability

Not stated: intended uses, limitations, acceptable use, input and output formats.
Not evidenced
Copyright policy published (Article 53(1)(c))

We found no published copyright policy. The Act asks for one to be in place, not necessarily published.

Not evidenced
Public summary of training content (Article 53(1)(d))

We found no published training-content summary from the provider.

Not evidenced
Authorised representative in the EU (Article 54)

We found no authorised representative named by the provider.

Not applicable
Adversarial testing (Article 55(1)(a))

Applies to models with systemic risk only.

Not applicable
Systemic risks assessed and mitigated (Article 55(1)(b))

Applies to models with systemic risk only.

Regulation (EU) 2024/1689, the EU AI Act · Source: General-Purpose AI Code of Practice, read 8 Nov 2026

Facthory’s EU AI Act score appears once a person has reviewed these checks. The articles we score against