LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Falcon 40B Instruct by tiiuae

 »  All LLMs  »  tiiuae  »  Falcon 40B Instruct   URL Share it on

  47356   

Falcon 40B Instruct is an open-source language model by tiiuae. Features: 40b LLM, VRAM: 83.6GB, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.21.

  Arxiv:1911.02150   Arxiv:2005.14165   Arxiv:2104.09864   Arxiv:2205.14135   Arxiv:2304.01196   Arxiv:2306.01116   Autotrain compatible   Custom code Dataset:tiiuae/falcon-refinedw...   En   Endpoints compatible   Falcon   Instruct   Pytorch   Region:us   Sharded
Model Card on HF 🤗: https://huggingface.co/tiiuae/falcon-40b-instruct 

Falcon 40B Instruct Benchmarks

MMLU Pro: 14.02
MUSR: 5.16
BBH: 17.22
IFEval: 24.54 vs 88 (so35)-72.1%
MATH Lvl 5: 1.96
LLME Score: 0.21128
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Falcon 40B Instruct (tiiuae/falcon-40b-instruct)
🌟 Advertise your project 🚀

Falcon 40B Instruct Parameters and Internals

Model Type 
Causal decoder-only
Use Cases 
Areas:
chatbot, instruction-following
Applications:
chat, instruction-based interactions
Primary Use Cases:
ready-to-use chat/instruct model
Limitations:
Model mostly trained on English data, may not generalize well to other languages
Considerations:
Develop guardrails and take precautions for production use.
Additional Notes 
Instruct model, not ideal for further finetuning. Optimized architecture for inference featuring FlashAttention and multiquery.
Supported Languages 
English (primary), French (secondary)
Training Details 
Data Sources:
Baize instruction dataset, RefinedWeb
Data Volume:
150M tokens from Baize mixed with 5% RefinedWeb
Methodology:
Finetuned on a mixture of chat data with 5% RefinedWeb
Context Length:
2048
Hardware Used:
64 A100 40GB GPUs on AWS SageMaker
Model Architecture:
Causal decoder-only with adaptations from GPT-3, including rotary embeddings, multiquery attention, FlashAttention, and a single layer norm with parallel attention/MLP
Input Output 
Accepted Modalities:
text
LLM NameFalcon 40B Instruct
Repository 🤗https://huggingface.co/tiiuae/falcon-40b-instruct 
Model Size40b
Required VRAM83.6 GB
Updated2025-10-09
Maintainertiiuae
Model Typefalcon
Instruction-BasedYes
Model Files  9.5 GB: 1-of-9   9.5 GB: 2-of-9   9.5 GB: 3-of-9   9.5 GB: 4-of-9   9.5 GB: 5-of-9   9.5 GB: 6-of-9   9.5 GB: 7-of-9   9.5 GB: 8-of-9   7.6 GB: 9-of-9
Supported Languagesen
Model ArchitectureFalconForCausalLM
Licenseapache-2.0
Model Max Length2048
Transformers Version4.26.0
Is Biased0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size65024
Torch Data Typebfloat16

Best Alternatives to Falcon 40B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
... Falcon 40B Instruct 4 Bit Bnb2K / 23.9 GB70
...Falcon 40B Instruct 4 Bit Gptq2K / 22.3 GB50

Rank the Falcon 40B Instruct Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

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  

What open-source LLMs or SLMs are you in search of? 52721 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer