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

Mpt 30B by mosaicml

 »  All LLMs  »  mosaicml  »  Mpt 30B   URL Share it on

  2931   

Mpt 30B is an open-source language model by mosaicml. Features: 30b LLM, VRAM: 60.1GB, License: apache-2.0, HF Score: 52.8, LLM Explorer Score: 0.14, Arc: 56, HellaSwag: 82.4, MMLU: 48, TruthfulQA: 38.4, WinoGrande: 74.9, GSM8K: 16.9.

  Arxiv:1909.08053   Arxiv:2010.04245   Arxiv:2108.12409   Arxiv:2205.14135   Arxiv:2302.06675   Arxiv:2302.13971   Autotrain compatible   Composer   Custom code   Dataset:allenai/c4   Dataset:allenai/s2orc Dataset:bigcode/the-stack-dedu...   Dataset:mc4 Dataset:togethercomputer/redpa...   Llm-foundry   Mosaicml   Mpt   Pytorch   Region:us   Sharded   Streamingdatasets
Model Card on HF 🤗: https://huggingface.co/mosaicml/mpt-30b 

Mpt 30B Benchmarks

ARC: 55.97 vs 96.7 (so35)-42.1%
HellaSwag: 82.42 vs 95.3 (gpt4)-13.5%
MMLU: 48 vs 88.3 (so35)-45.6%
TruthfulQA: 38.42 vs 59 (gpt4)-34.9%
WinoGrande: 74.9 vs 87.5 (gpt4)-14.4%
GSM8K: 16.91 vs 96.4 (so35)-82.5%
LLME Score: 0.14434
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Mpt 30B (mosaicml/mpt-30b)
🌟 Advertise your project 🚀

Mpt 30B Parameters and Internals

Model Type 
text generation
Training Details 
Data Sources:
allenai/c4, mc4, togethercomputer/RedPajama-Data-1T, bigcode/the-stack-dedup, allenai/s2orc
Data Volume:
1T tokens
Context Length:
8192
Hardware Used:
440 A100-40GB GPUs, 216 A100-40GB GPUs, 256 H100-80GB GPUs
Model Architecture:
Modified transformer architecture with FlashAttention, ALiBi, no biases
Input Output 
Performance Tips:
Use FlashAttention and ALiBi for faster training and inference.
LLM NameMpt 30B
Repository 🤗https://huggingface.co/mosaicml/mpt-30b 
Model Size30b
Required VRAM60.1 GB
Updated2025-09-23
Maintainermosaicml
Model Typempt
Model Files  9.8 GB: 1-of-7   9.9 GB: 2-of-7   9.9 GB: 3-of-7   9.9 GB: 4-of-7   9.9 GB: 5-of-7   9.9 GB: 6-of-7   0.8 GB: 7-of-7
Model ArchitectureMPTForCausalLM
Licenseapache-2.0
Model Max Length8192
Transformers Version4.28.1
Tokenizer ClassGPTNeoXTokenizer
Vocabulary Size50432
Torch Data Typebfloat16

Best Alternatives to Mpt 30B

Best Alternatives
Context / RAM
Downloads
Likes
Mpt 30B Chat0K / 60.1 GB952202
Mpt 30B Instruct0K / 60.1 GB2220103
Mpt 30B Orca Mini0K / 180.5 GB41
Mpt 30B V20K / 60.1 GB2410
Mpt 30B V30K / 60.1 GB142
...s Mpt 30B Gpt4 1p4 Five Epochs0K / 60.1 GB377
Mpt 30B Qlora Multi GPU0K /  GB21
Mpt 30B Peft Compatible0K / 60.1 GB118
...os Mpt 30B Gpt4 1p4 Six Epochs0K / 60.1 GB220
...t 30B Instruct Peft Compatible0K / 60.1 GB02
Note: green Score (e.g. "73.2") means that the model is better than mosaicml/mpt-30b.

Rank the Mpt 30B 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? 52392 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