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

Mistral7B PairRM SPPO Iter1 by UCLA-AGI

 »  All LLMs  »  UCLA-AGI  »  Mistral7B PairRM SPPO Iter1   URL Share it on

  23   

Mistral7B PairRM SPPO Iter1 is an open-source language model by UCLA-AGI. Features: 7.2b LLM, VRAM: 14.4GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.18.

  Arxiv:2405.00675   Conversational   Dataset:openbmb/ultrafeedback   Deploy:azure   En   Endpoints compatible   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/UCLA-AGI/Mistral7B-PairRM-SPPO-Iter1 

Mistral7B PairRM SPPO Iter1 Benchmarks

MMLU Pro: 18.84
GPQA: 4.47
MUSR: 8.3
BBH: 22.93
IFEval: 50.47 vs 88 (so35)-42.6%
MATH Lvl 5: 2.49
LLME Score: 0.18229
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Mistral7B PairRM SPPO Iter1 (UCLA-AGI/Mistral7B-PairRM-SPPO-Iter1)
🌟 Advertise your project 🚀

Mistral7B PairRM SPPO Iter1 Parameters and Internals

Model Type 
GPT-like, text generation
Additional Notes 
**This is the model reported in the paper**
Supported Languages 
en (primary)
Training Details 
Data Sources:
openbmb/UltraFeedback, snorkelai/Snorkel-Mistral-PairRM-DPO-Dataset
Data Volume:
synthetic
Methodology:
Self-Play Preference Optimization, synthetic responses, utilized the prompt sets, K=5 (generate 5 responses per iteration)
Model Architecture:
7B parameter GPT-like
LLM NameMistral7B PairRM SPPO Iter1
Repository 🤗https://huggingface.co/UCLA-AGI/Mistral7B-PairRM-SPPO-Iter1 
Model Size7.2b
Required VRAM14.4 GB
Updated2026-04-17
MaintainerUCLA-AGI
Model Typemistral
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   4.5 GB: 3-of-3
Supported Languagesen
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to Mistral7B PairRM SPPO Iter1

Best Alternatives
Context / RAM
Downloads
Likes
...Web AI HumanAI 012 INSTRUCT XA512K / 14.4 GB90
...Web AI HumanAI 012 INSTRUCT IA512K / 14.4 GB80
...Web AI HumanAI 011 INSTRUCT ML512K / 14.4 GB70
14512K / 14.5 GB3190
SpydazWeb HumanAI M2512K /  GB380
...dazWeb AI HumanAI 011 INSTRUCT512K / 14.4 GB141
SpydazWeb AI HumanAI 009 CHAT512K / 14.4 GB60
SpydazWeb AI HumanAI 007512K / 14.4 GB50
...Web AI HumanAI 012 INSTRUCT MX512K / 14.4 GB80
SpydazWeb AI HumanAI 010 CHAT512K / 14.4 GB70
Note: green Score (e.g. "73.2") means that the model is better than UCLA-AGI/Mistral7B-PairRM-SPPO-Iter1.

Rank the Mistral7B PairRM SPPO Iter1 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? 53053 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