LLM EXPLORER 59,420 MODELS INDEXED

Orthrus 12B V0.8 by Pyroserenus

By Pyroserenus · 4 downloads
HF Repo ↗ Copy URL

Orthrus 12B V0.8 is an open-source language model by Pyroserenus. Features: 12b LLM, VRAM: 24.5GB, License: cc-by-nc-4.0, Instruction-Based, Merged, LLM Explorer Score: 0.14.

  Merged Model   Arxiv:2306.01708 Base model:anthracite-org/magn... Base model:intervitensinc/mist... Base model:unsloth/mistral-nem...   Endpoints compatible   Instruct   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Pyroserenus/Orthrus-12b-v0.8 

Orthrus 12B V0.8 Parameters and Internals

Additional Notes 
This model was meant as a component model for a larger merge but is usable as is.
Training Details 
Methodology:
[TIES](https://arxiv.org/abs/2306.01708) merge method using [IntervitensInc/Mistral-Nemo-Base-2407-chatml](https://huggingface.co/IntervitensInc/Mistral-Nemo-Base-2407-chatml) as a base.
Input Output 
Input Format:
ChatML preferred
Performance Tips:
Tested using 0.75 temp, 0.10 MinP, 0.30 Smoothing Factor.
LLM NameOrthrus 12B V0.8
Repository 🤗https://huggingface.co/Pyroserenus/Orthrus-12b-v0.8 
Base Model(s)  Magnum 12B V2   Mistral Nemo Instruct 2407   Mistral Nemo Base 2407 Chatml   anthracite-org/magnum-12b-v2   unsloth/Mistral-Nemo-Instruct-2407   IntervitensInc/Mistral-Nemo-Base-2407-chatml
Merged ModelYes
Model Size12b
Required VRAM24.5 GB
Updated2026-07-13
MaintainerPyroserenus
Instruction-BasedYes
Model Files  4.9 GB: 1-of-5   4.9 GB: 2-of-5   4.9 GB: 3-of-5   4.9 GB: 4-of-5   4.9 GB: 5-of-5
Model ArchitectureAutoModelForCausalLM
Licensecc-by-nc-4.0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size131072

Best Alternatives to Orthrus 12B V0.8

Best Alternatives
Context / RAM
Downloads
Likes
Ultra Instruct 12B0K / 24.5 GB244
CleverBoi Nemo 12B V20K / 0.2 GB3404
Note: green Score (e.g. "73.2") means that the model is better than Pyroserenus/Orthrus-12b-v0.8.