LLM EXPLORER 59,420 MODELS INDEXED

Aguila 7B by projecte-aina

By projecte-aina · 33 downloads
HF Repo ↗ Copy URL

Aguila 7B is an open-source language model by projecte-aina. Features: 7b LLM, VRAM: 13.7GB, Context: 2K, LLM Explorer Score: 0.08.

  Aguila   Ca   Catalan   Custom code   En   Endpoints compatible   Es   Falcon   Model-index   Pytorch   Refinedwebmodel   Region:us   Safetensors   Sharded   Spanish   Tensorflow
Model Card on HF 🤗: https://huggingface.co/projecte-aina/aguila-7b 

Aguila 7B Parameters and Internals

Model Type 
causal language model, text generation
Use Cases 
Areas:
research, commercial applications
Primary Use Cases:
causal language modeling, text-generation tasks
Limitations:
Limited to training data biases; No bias and toxicity estimates currently available.
Considerations:
Model is trained on data that may contain biases and should be used with caution.
Additional Notes 
Small amount of English data was retained to prevent catastrophic forgetting.
Supported Languages 
en (High), es (High), ca (High)
Training Details 
Data Sources:
Wikipedia, C4_es, Biomedical, Legal, Gutenberg, C4_ca, RacoCatalà Noticias, RacoCatalà Forums, CaWaC, Vilaweb
Data Volume:
26B tokens
Methodology:
Adapted by swapping the tokenizer and adjusting the embedding layer.
Training Time:
320 hours
Hardware Used:
8 NVIDIA H100 GPUs with 80GB RAM
Model Architecture:
Byte-Pair Encoding (BPE) tokenizer with 50,257 tokens.
Responsible Ai Considerations 
Fairness:
No measures have been taken to estimate bias and toxicity at the time of submission.
Transparency:
Model is provided with documentation of its creation and intended use.
Accountability:
Accountability lies with the users deploying the model.
Mitigation Strategies:
The model users should aim to mitigate risks associated with bias and toxicity.
Input Output 
Input Format:
Text prompts
Accepted Modalities:
text
Output Format:
Text generation
Release Notes 
LLM NameAguila 7B
Repository 🤗https://huggingface.co/projecte-aina/aguila-7b 
Model Size7b
Required VRAM13.7 GB
Updated2026-08-02
Maintainerprojecte-aina
Model TypeRefinedWebModel
Model Files  1.9 GB: 1-of-8   1.9 GB: 2-of-8   2.0 GB: 3-of-8   1.9 GB: 4-of-8   1.9 GB: 5-of-8   2.0 GB: 6-of-8   1.9 GB: 7-of-8   0.2 GB: 8-of-8   0.0 GB
Supported Languagesen es ca
Model ArchitectureRWForCausalLM
Context Length2048
Model Max Length2048
Transformers Version4.35.2
Is Biased0
Tokenizer ClassGPT2Tokenizer
Beginning of Sentence Token<|endoftext|>
End of Sentence Token<|endoftext|>
Unk Token<|endoftext|>
Vocabulary Size50257
Torch Data Typefloat16
Errorsreplace

Best Alternatives to Aguila 7B

Best Alternatives
Context / RAM
Downloads
Likes
Aguila Falcon InstruCATPlus2K / 13.7 GB100
Aguila Falcon Instrucat2K / 13.7 GB250
Falcon Aguila Meteocatv22K / 13.7 GB60
Falcon Aguila Meteocat2K / 13.7 GB220
Testing6000v20K / 15.1 GB60
Ct2 Int8 Falcon 7B Instruct0K /  GB50
Falcon No Parallel Attn Test0K / 0 GB177050
...ce Falcon 7b Sharded Quantized0K / 13.8 GB113
...ce Falcon 7b Sharded Quantized0K / 13.8 GB111
Falcon 7b Python Instructions0K / 13.8 GB51
Note: green Score (e.g. "73.2") means that the model is better than projecte-aina/aguila-7b.