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

Open Llama 3B V2 by openlm-research

 »  All LLMs  »  openlm-research  »  Open Llama 3B V2   URL Share it on

  30478   
  Autotrain compatible   Dataset:bigcode/starcoderdata Dataset:tiiuae/falcon-refinedw... Dataset:togethercomputer/redpa...   Endpoints compatible   Llama   Pytorch   Region:us
Model Card on HF 🤗: https://huggingface.co/openlm-research/open_llama_3b_v2 

Open Llama 3b V2 Benchmarks

MMLU Pro: 1.06
GPQA: 3.58
MUSR: 4.69
BBH: 2.81
IFEval: 18.72 vs 88 (so35)-78.7%
ARC: 46.6 vs 96.7 (so35)-51.8%
HellaSwag: 68.2 vs 95.3 (gpt4)-28.4%
MMLU: 27.3 vs 88.3 (so35)-69.1%
TruthfulQA: 34.78 vs 59 (gpt4)-41.1%
WinoGrande: 67.01 vs 87.5 (gpt4)-23.4%
GSM8K: 3.1 vs 96.4 (so35)-96.8%
OpenCompas: 34.6 vs 73.3 (gpt4)-52.8%
Exam: 31 vs 77.2 (gpt4)-59.8%
Language: 33.2 vs 62 (gpt4)-46.5%
Knowledge: 42.1 vs 73.5 (gpt4)-42.7%
Understand: 33.2 vs 70 (gpt4)-52.6%
Reasoning: 34.4 vs 74.4 (gpt4)-53.8%
LLME Score: 0.22922
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Open Llama 3B V2 (openlm-research/open_llama_3b_v2)

Open Llama 3B V2 Parameters and Internals

Model Type 
large language model
Additional Notes 
Please note that it is advised to avoid using the Hugging Face fast tokenizer for now, as it sometimes gives incorrect tokenizations. This can be avoided by using 'use_fast=False'.
Training Details 
Data Sources:
tiiuae/falcon-refinedweb, bigcode/starcoderdata, togethercomputer/RedPajama-Data-1T
Data Volume:
1 trillion tokens
Methodology:
Pre-trained with open datasets rather than the original LLaMA dataset, using the EasyLM framework.
Hardware Used:
cloud TPU-v4s
LLM NameOpen Llama 3b V2
Repository 🤗https://huggingface.co/openlm-research/open_llama_3b_v2 
Model Size3b
Required VRAM6.8 GB
Updated2025-03-12
Maintaineropenlm-research
Model Typellama
Model Files  6.8 GB
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.31.0.dev0
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Quantized Models of the Open Llama 3B V2

Model
Likes
Downloads
VRAM
...3b V2 Python Instruct 0.1 4bit01552 GB

Best Alternatives to Open Llama 3B V2

Best Alternatives
Context / RAM
Downloads
Likes
Llama 3.2 3B Instruct128K / 6.5 GB30134691190
Hermes 3 Llama 3.2 3B128K / 6.5 GB351400147
Llama 3.2 3B128K / 6.5 GB322803527
DeepSeek R1 Distill Llama 3B128K / 6.5 GB7369
Llama 3.2 3B ToxicKod128K / 6.4 GB322
ReasoningCore 3B RE1 V2128K / 6.5 GB2730
...penReasoner Llama 3.2 3B Rs1.0128K / 6.5 GB2271
Zeitgeist 3B V1128K / 6.5 GB1055
Llama 3.2 3B Khelavaster128K / 7.2 GB111
Eximius Persona 5B128K / 11.6 GB4534
Note: green Score (e.g. "73.2") means that the model is better than openlm-research/open_llama_3b_v2.

Rank the Open Llama 3B V2 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? 44949 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 v20241227
  Support LLM Explorer