LLM EXPLORER 58,425 MODELS INDEXED

Tosilos 128B by nesilabs

By nesilabs · 60 downloads
HF Repo ↗ Copy URL

Tosilos 128B is an open-source language model by nesilabs. Features: 128b LLM, VRAM: 255.4GB, License: other, Quantized, LLM Explorer Score: 0.44, ELO: 1387.

Base model:mistralai/mistral-m... Base model:quantized:mistralai...   Bugbounty   Conversational   Cybersecurity   Endpoints compatible   European   Exploit   Gguf   Mistral   Mistral3   Pentest   Q3   Qlora   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/nesilabs/tosilos-128b 

Tosilos 128B Benchmarks

LMSys ELO 1387 vs 1272 (so35) 9%
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

Tosilos 128B Parameters and Internals

LLM NameTosilos 128B
Repository 🤗https://huggingface.co/nesilabs/tosilos-128b 
Base Model(s)  mistralai/Mistral-Medium-3.5-128B   mistralai/Mistral-Medium-3.5-128B
Model Size128b
Required VRAM255.4 GB
Updated2026-08-14
Maintainernesilabs
Model Typemistral3
Model Files  90.9 GB: 1-of-3   99.7 GB: 2-of-3   64.8 GB: 3-of-3   60.6 GB   74.9 GB
GGUF QuantizationYes
Quantization Typegguf|q3|q4_k
Model ArchitectureMistral3ForConditionalGeneration
Licenseother
Transformers Version5.6.0.dev0
Tokenizer ClassTokenizersBackend
Padding Token<pad>
Torch Data Typebfloat16

Best Alternatives to Tosilos 128B

Best Alternatives
Context / RAM
Downloads
Likes
Mistral Medium 3.5 128B GGUF0K / 5.4 GB2046983
Mistral Medium 3.5 128B BF160K / 254.9 GB33720
Note: green Score (e.g. "73.2") means that the model is better than nesilabs/tosilos-128b.