LLM EXPLORER 59,420 MODELS INDEXED

TinyMistral 248M by Locutusque

By Locutusque · 191 downloads
HF Repo ↗ Copy URL

TinyMistral 248M is an open-source language model by Locutusque. Features: 248m LLM, VRAM: 1GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.1.

  Dataset:jeankaddour/minipile   Dataset:skylion007/openwebtext   Deploy:azure   En   Endpoints compatible   Mistral   Pytorch   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/Locutusque/TinyMistral-248M 

TinyMistral 248M Parameters and Internals

Model Type 
text-generation
Use Cases 
Primary Use Cases:
fine-tuning on a downstream task
Additional Notes 
This model aims to prove trillion-scale datasets are not necessary for language model pretraining.
Supported Languages 
en (fluent)
Training Details 
Data Sources:
Skylion007/openwebtext, JeanKaddour/minipile
Data Volume:
7,488,000 examples
Context Length:
32768
Hardware Used:
single GPU (Titan V)
Input Output 
Accepted Modalities:
text
Output Format:
text-generation
LLM NameTinyMistral 248M
Repository 🤗https://huggingface.co/Locutusque/TinyMistral-248M 
Model Size248m
Required VRAM1 GB
Updated2026-08-01
MaintainerLocutusque
Model Typemistral
Model Files  1.0 GB   1.0 GB
Supported Languagesen
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.35.0
Tokenizer ClassLlamaTokenizer
Padding Token[PAD]
Vocabulary Size32005
Torch Data Typefloat16

Best Alternatives to TinyMistral 248M

Best Alternatives
Context / RAM
Downloads
Likes
TinyMistral 248M V2.532K / 1 GB22627
TinyMistral 248M V232K / 1 GB109617
TinyMistral 248M V2.5 Instruct32K / 1 GB211
...al V2.5 MiniPile Guidelines E132K / 0.6 GB132
TinyMistral V2 Test132K / 1 GB181
...adrin TinyMistral248M Instruct32K / 0.5 GB1087
TinyMistral 248M Instruct32K / 1 GB6912
TinyMistral 248M 8bits32K / 0.3 GB211
Tinymistv132K / 0.5 GB170
TinyMistral Haiku32K / 1 GB100
Note: green Score (e.g. "73.2") means that the model is better than Locutusque/TinyMistral-248M.