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

Mahou 1.2a Mistral 7B by flammenai

 »  All LLMs  »  flammenai  »  Mahou 1.2a Mistral 7B   URL Share it on

  39   
  Autotrain compatible Base model:flammenai/flammen27... Base model:quantized:flammenai... Dataset:flammenai/flamemix-dpo... Dataset:flammenai/grill-prepro... Dataset:flammenai/grill-prepro...   Endpoints compatible   Gguf   Mistral   Q3   Quantized   Region:us   Safetensors   Sharded   Tensorflow

Mahou 1.2a Mistral 7B Benchmarks

MMLU Pro: 24.04
GPQA: 2.91
MUSR: 6.97
BBH: 31.17
IFEval: 45.37 vs 88 (so35)-48.4%
ARC: 68.77 vs 96.7 (so35)-28.9%
HellaSwag: 88.14 vs 95.3 (gpt4)-7.5%
MMLU: 64.73 vs 88.3 (so35)-26.7%
TruthfulQA: 68.71 vs 59 (gpt4)16.5%
WinoGrande: 80.11 vs 87.5 (gpt4)-8.4%
GSM8K: 63.61 vs 96.4 (so35)-34%
MATH Lvl 5: 5.82
LLME Score: 0.27857
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

Mahou 1.2a Mistral 7B Parameters and Internals

Model Type 
conversational, roleplay, chat
Additional Notes 
Fine-tuned from flammen.ai conversational data to improve comprehension and coherence. Uses ChatML and roleplay formats.
Training Details 
Data Sources:
flammenai/FlameMix-DPO-v1, flammenai/Grill-preprod-v1_chatML, flammenai/Grill-preprod-v2_chatML
Methodology:
Fine-tuned with Direct Preference Optimization (DPO) using LoRA configurations.
Hardware Used:
A100 on Google Colab
Model Architecture:
LoRA configuration with specific modules targeted.
Input Output 
Input Format:
ChatML format
LLM NameMahou 1.2a Mistral 7B
Repository 🤗https://huggingface.co/flammenai/Mahou-1.2a-mistral-7B 
Base Model(s)  Flammen27 Mistral 7B   flammenai/flammen27-mistral-7B
Model Size7b
Required VRAM14.4 GB
Updated2024-11-21
Maintainerflammenai
Model Typemistral
Model Files  3.5 GB   4.4 GB   5.1 GB   4.9 GB: 1-of-3   5.0 GB: 2-of-3   4.5 GB: 3-of-3
GGUF QuantizationYes
Quantization Typeq3|gguf|q4_k|q5_k
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.40.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size32000
Torch Data Typebfloat16
Mahou 1.2a Mistral 7B (flammenai/Mahou-1.2a-mistral-7B)

Quantized Models of the Mahou 1.2a Mistral 7B

Model
Likes
Downloads
VRAM
Mahou 1.2a Mistral 7B AWQ1134 GB

Best Alternatives to Mahou 1.2a Mistral 7B

Best Alternatives
Context / RAM
Downloads
Likes
MegaBeam Mistral 7B 300K Gguf282K / 5 GB213
Mistral 7B Instruct V0.2.gguf32K / 14.5 GB103
Mistral 7B Instruct V0.2 GGUF32K / 4.4 GB532
...andle Mistral 7B Instruct V0.232K / 14.4 GB410
BioMistralMerged32K / 14.4 GB39030
Tamil Mistral 7B Instruct V0.132K / 14.8 GB55113
Mistroll 7B V2.232K / 14.4 GB310812
... Noromaid NSFW Mistral 7B GGUF32K / 4.1 GB1399630
Dgtkqabot Mistral 7B V332K / 14.5 GB4350
...stral 7B V0.3 Chinese Chat F1632K / 14.5 GB6341
Note: green Score (e.g. "73.2") means that the model is better than flammenai/Mahou-1.2a-mistral-7B.

Rank the Mahou 1.2a Mistral 7B 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? 38149 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.
Release v20241110
  Support LLM Explorer