LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

MistralTrix V1 by CultriX

 »  All LLMs  »  CultriX  »  MistralTrix V1   URL Share it on

  661   

MistralTrix V1 is an open-source language model by CultriX. Features: 9b LLM, VRAM: 17.9GB, Context: 32K, License: apache-2.0, Merged, HF Score: 73.4, LLM Explorer Score: 0.14, Arc: 72.3, HellaSwag: 88.3, MMLU: 65.2, TruthfulQA: 70.7, WinoGrande: 81, GSM8K: 62.8.

  Merged Model   En   Endpoints compatible   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/CultriX/MistralTrix-v1 

MistralTrix V1 Benchmarks

ARC: 72.27 vs 96.7 (so35)-25.3%
HellaSwag: 88.33 vs 95.3 (gpt4)-7.3%
MMLU: 65.24 vs 88.3 (so35)-26.1%
TruthfulQA: 70.73 vs 59 (gpt4)19.9%
WinoGrande: 80.98 vs 87.5 (gpt4)-7.5%
GSM8K: 62.77 vs 96.4 (so35)-34.9%
LLME Score: 0.13653
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
MistralTrix V1 (CultriX/MistralTrix-v1)
🌟 Advertise your project 🚀

MistralTrix V1 Parameters and Internals

Model Type 
text-generation
Additional Notes 
Model has been further fine-tuned to surpass original benchmarks and is inspired by RLHF processes described by Intel.
Supported Languages 
en (proficient)
Training Details 
Data Sources:
Intel's dataset for neural-chat-7b-v3-1
Methodology:
Fine-tuning with Direct Preference Optimization (DPO) and applying ChatML template.
Context Length:
1024
Training Time:
less than a few hours
Hardware Used:
Google Colab A-1000 GPU with 40GB VRAM
LLM NameMistralTrix V1
Repository 🤗https://huggingface.co/CultriX/MistralTrix-v1 
Merged ModelYes
Model Size9b
Required VRAM17.9 GB
Updated2026-04-01
MaintainerCultriX
Model Typemistral
Model Files  4.9 GB: 1-of-4   5.0 GB: 2-of-4   4.9 GB: 3-of-4   3.1 GB: 4-of-4
Supported Languagesen
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.35.2
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to MistralTrix V1

Best Alternatives
Context / RAM
Downloads
Likes
Peagle 9B32K / 18 GB128140
Peagle 9B32K / 18 GB124370
MetaMath Mistral Pro32K / 17.9 GB55
...onarch Neurotic 9B Passthrough32K / 17.9 GB80
GML Mistral Merged V132K / 18 GB7211
Mistral 9B Instruct32K / 35.9 GB70
Garbage 9B32K / 17.9 GB933
Bepis 9B32K / 18 GB256
Mistral 9B Instruct V0.232K / 17.9 GB53
Notus TheTop 7B Passthrough32K / 18 GB50
Note: green Score (e.g. "73.2") means that the model is better than CultriX/MistralTrix-v1.

Rank the MistralTrix V1 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! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 52721 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer