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

Airoboros L2 70B 3.1.2 by jondurbin

 »  All LLMs  »  jondurbin  »  Airoboros L2 70B 3.1.2   URL Share it on

  1798   
  Autotrain compatible   Conversational Dataset:jondurbin/airoboros-3....   Endpoints compatible   Llama   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/jondurbin/airoboros-l2-70b-3.1.2 

Airoboros L2 70B 3.1.2 Benchmarks

ARC: 70.14 vs 96.7 (so35)-27.5%
HellaSwag: 86.88 vs 95.3 (gpt4)-8.8%
MMLU: 69.72 vs 88.3 (so35)-21%
TruthfulQA: 59.19 vs 59 (gpt4)0.3%
WinoGrande: 83.11 vs 87.5 (gpt4)-5%
GSM8K: 49.43 vs 96.4 (so35)-48.7%
LLME Score: 0.14384
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Airoboros L2 70B 3.1.2 (jondurbin/airoboros-l2-70b-3.1.2)
🌟 Advertise your project 🚀

Airoboros L2 70B 3.1.2 Parameters and Internals

Model Type 
instruction following
Use Cases 
Areas:
Research, Instruction following
Applications:
Instruction following tasks, Math problem solving using MathJSON
Primary Use Cases:
Mathematical problem solving, Contextual Q&A, Summarization, Multi-step instruction following
Limitations:
Not suitable for casual chat/roleplay
Considerations:
Use a low temperature for math and context obedient question answering.
Additional Notes 
Configured for MathJSON and closed-context question answering. Use special prompts for chain-of-thought reasoning and execution planning.
Training Details 
Data Sources:
airoboros-3.1 dataset, airoboros-3.0 dataset
Methodology:
Self-instruct with synthetic data generation using OpenAI API calls to GPT-4
Model Architecture:
Llama2 chat format
Input Output 
Input Format:
Llama-2 chat format
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Use a low temperature for better results. Use specific structured prompts for different tasks.
LLM NameAiroboros L2 70B 3.1.2
Repository 🤗https://huggingface.co/jondurbin/airoboros-l2-70b-3.1.2 
Model Size70b
Required VRAM138 GB
Updated2025-09-23
Maintainerjondurbin
Model Typellama
Model Files  4.0 GB: 1-of-36   3.7 GB: 2-of-36   3.9 GB: 3-of-36   3.9 GB: 4-of-36   3.9 GB: 5-of-36   3.7 GB: 6-of-36   3.9 GB: 7-of-36   3.9 GB: 8-of-36   3.9 GB: 9-of-36   3.7 GB: 10-of-36   3.9 GB: 11-of-36   3.9 GB: 12-of-36   3.9 GB: 13-of-36   3.7 GB: 14-of-36   3.9 GB: 15-of-36   3.9 GB: 16-of-36   3.9 GB: 17-of-36   3.7 GB: 18-of-36   3.9 GB: 19-of-36   3.9 GB: 20-of-36   3.9 GB: 21-of-36   3.7 GB: 22-of-36   3.9 GB: 23-of-36   3.9 GB: 24-of-36   3.9 GB: 25-of-36   3.7 GB: 26-of-36   3.9 GB: 27-of-36   3.9 GB: 28-of-36   3.9 GB: 29-of-36   3.7 GB: 30-of-36   3.9 GB: 31-of-36   3.9 GB: 32-of-36   3.9 GB: 33-of-36   3.7 GB: 34-of-36   3.9 GB: 35-of-36   3.2 GB: 36-of-36
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length4096
Model Max Length4096
Transformers Version4.34.1
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typebfloat16

Quantized Models of the Airoboros L2 70B 3.1.2

Model
Likes
Downloads
VRAM
Airoboros L2 70B 3.1.2 GGUF1023029 GB
Airoboros L2 70B 3.1.2 GPTQ8635 GB
Airoboros L2 70B 3.1.2 AWQ3936 GB

Best Alternatives to Airoboros L2 70B 3.1.2

Best Alternatives
Context / RAM
Downloads
Likes
... Chat 1048K Chinese Llama3 70B1024K / 141.9 GB98265
... Chat 1048K Chinese Llama3 70B1024K / 141.9 GB90695
... 3 70B Instruct Gradient 1048K1024K / 141.9 GB13122
Llama3 Function Calling 1048K1024K / 141.9 GB61
...a 3 70B Instruct Gradient 524K512K / 141.9 GB1023
...a 3 70B Instruct Gradient 262K256K / 141.9 GB11456
...ama 3 70B Arimas Story RP V2.0256K / 141.1 GB263
...ama 3 70B Arimas Story RP V1.6256K / 141.2 GB130
...ama 3 70B Arimas Story RP V1.5256K / 141.2 GB463
Yi 70B 200K RPMerge Franken195K / 142.4 GB21
Note: green Score (e.g. "73.2") means that the model is better than jondurbin/airoboros-l2-70b-3.1.2.

Rank the Airoboros L2 70B 3.1.2 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? 51544 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 v20241124
  Support LLM Explorer