TinyMistral 248M V2 is an open-source language model by Locutusque. Features: 248m LLM, VRAM: 1GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.17, Arc: 21.3, HellaSwag: 26.6, MMLU: 23.4.
Dataset:locutusque/tm-data Dataset:skylion007/openwebtext En Endpoints compatible Mistral Pytorch Region:us Safetensors
Model Card on HF 🤗: https://huggingface.co/Locutusque/TinyMistral-248M-v2
TinyMistral 248M V2 Benchmarks
ARC: 21.25 vs 96.7 (so35)-78%
HellaSwag: 26.56 vs 95.3 (gpt4)-72.1%
MMLU: 23.39 vs 88.3 (so35)-73.5%
LLME Score: 0.16697
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
TinyMistral 248M V2 Parameters and Internals
| Model Type |
| |||||||||
| Supported Languages |
| |||||||||
| Training Details |
|
| LLM Name | TinyMistral 248M V2 |
| Repository 🤗 | https://huggingface.co/Locutusque/TinyMistral-248M-v2 |
| Model Size | 248m |
| Required VRAM | 1 GB |
| Updated | 2026-05-01 |
| Maintainer | Locutusque |
| Model Type | mistral |
| Model Files | |
| Supported Languages | en |
| Model Architecture | MistralForCausalLM |
| License | apache-2.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.36.2 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 32005 |
| Torch Data Type | float32 |
Best Alternatives to TinyMistral 248M V2
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| TinyMistral 248M V2.5 | 32K / 1 GB | 205 | 27 |
| ...adrin TinyMistral248M Instruct | 32K / 0.5 GB | 2127 | 7 |
| TinyMistral 248M V2.5 Instruct | 32K / 1 GB | 2 | 11 |
| ...al V2.5 MiniPile Guidelines E1 | 32K / 0.6 GB | 8 | 2 |
| TinyMistral 248M Instruct | 32K / 1 GB | 851 | 12 |
| TinyMistral V2 Test1 | 32K / 1 GB | 18 | 1 |
| TinyMistral 248M 8bits | 32K / 0.3 GB | 35 | 1 |
| Tinymistv1 | 32K / 0.5 GB | 17 | 0 |
| TinyMistral Haiku | 32K / 1 GB | 10 | 0 |
| ...istral 248M V2.5 Instruct Orpo | 32K / 0.5 GB | 8 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than Locutusque/TinyMistral-248M-v2 .
Rank the TinyMistral 248M V2 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! 🌟
What open-source LLMs or SLMs are you in search of? 53999 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a