LLM EXPLORER 66,613 MODELS INDEXED

Ct2fast Falcon 40B Instruct by michaelfeil

By michaelfeil · 2 downloads
HF Repo ↗ Copy URL

Ct2fast Falcon 40B Instruct is an open-source language model by michaelfeil. Features: 40b LLM, VRAM: 41.3GB, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.07.

  Arxiv:1911.02150   Arxiv:2005.14165   Arxiv:2104.09864   Arxiv:2205.14135   Arxiv:2304.01196   Arxiv:2306.01116   Ctranslate2 Dataset:tiiuae/falcon-refinedw...   En   Float16   Instruct   Int8   Region:us
Model Card on HF 🤗: https://huggingface.co/michaelfeil/ct2fast-falcon-40b-instruct 

Ct2fast Falcon 40B Instruct Parameters and Internals

Model Type 
Causal decoder-only
Use Cases 
Areas:
Research, Commercial applications
Limitations:
Mostly trained on English data, May carry online stereotypes and biases
Considerations:
Develop guardrails and take appropriate precautions for production use
Additional Notes 
This is a quantized version. License conditions are identical to the original model's HuggingFace repository.
Supported Languages 
English (NLP), French (NLP)
Training Details 
Data Sources:
Baize, Falcon RefinedWeb
Data Volume:
150M tokens
Methodology:
Finetuned on a mixture of Baize chat dataset and 5% of RefinedWeb data
Context Length:
2048
Hardware Used:
64 A100 40GB GPUs in P4d instances
Model Architecture:
Causal decoder-only with GPT-3 inspired architecture
LLM NameCt2fast Falcon 40B Instruct
Repository 🤗https://huggingface.co/michaelfeil/ct2fast-falcon-40b-instruct 
Model Size40b
Required VRAM41.3 GB
Updated2026-08-04
Maintainermichaelfeil
Instruction-BasedYes
Model Files  41.3 GB
Supported Languagesen
Model ArchitectureAutoModel
Licenseapache-2.0
Model Max Length2048
Tokenizer ClassPreTrainedTokenizerFast

Best Alternatives to Ct2fast Falcon 40B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
Falcon 40B Instruct GGML0K / 13.7 GB1058
Note: green Score (e.g. "73.2") means that the model is better than michaelfeil/ct2fast-falcon-40b-instruct.