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

Hermes 2 Pro Mistral 7B by NousResearch

 »  All LLMs  »  NousResearch  »  Hermes 2 Pro Mistral 7B   URL Share it on

  13505   
  Autotrain compatible Base model:finetune:mistralai/... Base model:mistralai/mistral-7...   Chatml   Conversational   Dataset:teknium/openhermes-2.5   Distillation   Dpo   En   Endpoints compatible   Finetuned   Function calling   Gpt4   Instruct   Json mode   Mistral   Region:us   Rlhf   Safetensors   Sharded   Synthetic data   Tensorflow
Model Card on HF 🤗: https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B 

Hermes 2 Pro Mistral 7B Benchmarks

MMLU Pro: 21.63
GPQA: 3.13
MUSR: 14.13
BBH: 29.43
IFEval: 56.68 vs 88 (so35)-35.6%
ARC: 64.16 vs 96.7 (so35)-33.7%
HellaSwag: 82.73 vs 95.3 (gpt4)-13.2%
MMLU: 62.21 vs 88.3 (so35)-29.5%
TruthfulQA: 58.99 vs 59 (gpt4)-0%
WinoGrande: 75.61 vs 87.5 (gpt4)-13.6%
GSM8K: 60.42 vs 96.4 (so35)-37.3%
MATH Lvl 5: 5.21
LLME Score: 0.37299
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Hermes 2 Pro Mistral 7B (NousResearch/Hermes-2-Pro-Mistral-7B)

Hermes 2 Pro Mistral 7B Parameters and Internals

Model Type 
text generation, function calling, structured outputs
Use Cases 
Areas:
general task, conversation, function calling
Additional Notes 
Developed in collaboration with Fireworks.AI and @interstellarninja.
Training Details 
Data Sources:
OpenHermes 2.5 Dataset, Function Calling and JSON Mode dataset developed in-house
Methodology:
Special system prompt and multi-turn function calling structure with ChatML.
Input Output 
Input Format:
ChatML - special tokens for beginning and end of turns, roles.
Accepted Modalities:
text
Output Format:
text, JSON structured outputs.
Performance Tips:
Set 'add_generation_prompt=True' when tokenizing messages for generation using 'apply_chat_template()'.
Release Notes 
Version:
Hermes-2-Pro-Mistral-7B
Notes:
Enhanced with function calling capabilities and JSON mode responses.
LLM NameHermes 2 Pro Mistral 7B
Repository 🤗https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B 
Base Model(s)  mistralai/Mistral-7B-v0.1   mistralai/Mistral-7B-v0.1
Model Size7b
Required VRAM14.5 GB
Updated2024-12-22
MaintainerNousResearch
Model Typemistral
Model Files  4.0 GB: 1-of-4   3.9 GB: 2-of-4   3.9 GB: 3-of-4   2.7 GB: 4-of-4
Supported Languagesen
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.38.2
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size32032
Torch Data Typebfloat16

Quantized Models of the Hermes 2 Pro Mistral 7B

Model
Likes
Downloads
VRAM
...Mistral 7B Metropole 4bit Gguf0154 GB
Hermes 2 Pro Mistral 7B AWQ2554 GB

Best Alternatives to Hermes 2 Pro Mistral 7B

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB26689
MegaBeam Mistral 7B 512K512K / 14.4 GB715646
SpydazWeb AI HumanAI RP512K / 14.4 GB331
SpydazWeb AI HumanAI 002512K / 14.4 GB201
...daz Web AI ChatML 512K Project512K / 14.5 GB120
MegaBeam Mistral 7B 300K282K / 14.4 GB328215
Hebrew Mistral 7B 200K256K / 30 GB319315
Astral 256K 7B V2250K / 14.4 GB190
Astral 256K 7B250K / 14.4 GB160
Boptruth Agatha 7B128K / 14.4 GB4750

Rank the Hermes 2 Pro 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? 40066 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 v20241217
  Support LLM Explorer