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

Quokka 2.7B by Corianas

 »  All LLMs  »  Corianas  »  Quokka 2.7B   URL Share it on

  848   

Quokka 2.7B is an open-source language model by Corianas. Features: 2.7b LLM, VRAM: 5.4GB, License: apache-2.0, HF Score: 33.3, LLM Explorer Score: 0.13, Arc: 31.1, HellaSwag: 47.7, MMLU: 24.8, TruthfulQA: 40.1, WinoGrande: 55.5, GSM8K: 0.4.

  Arxiv:1910.09700   Dataset:guanaco/guanaco   Dataset:the pile   En   Endpoints compatible   Gpt2   Pytorch   Region:us   Safetensors   Tensorboard
Model Card on HF 🤗: https://huggingface.co/Corianas/Quokka_2.7b 

Quokka 2.7B Benchmarks

MMLU Pro: 1.61
GPQA: 0.78
MUSR: 6.09
BBH: 3.17
IFEval: 17.49 vs 88 (so35)-80.1%
ARC: 31.06 vs 96.7 (so35)-67.9%
HellaSwag: 47.72 vs 95.3 (gpt4)-49.9%
MMLU: 24.8 vs 88.3 (so35)-71.9%
TruthfulQA: 40.14 vs 59 (gpt4)-32%
WinoGrande: 55.49 vs 87.5 (gpt4)-36.6%
GSM8K: 0.38 vs 96.4 (so35)-99.6%
MATH Lvl 5: 0.83
LLME Score: 0.13106
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Quokka 2.7B (Corianas/Quokka_2.7b)
🌟 Advertise your project 🚀

Quokka 2.7B Parameters and Internals

Use Cases 
Primary Use Cases:
Chatbot
Limitations:
Not suitable for accurate information, Do not use for medical or financial advice
Additional Notes 
The model is a fine-tuned version of the existing Cerebras model, trained to follow instructions with a focus on conversational output. Its factual reliability is low.
Supported Languages 
en (unknown)
Training Details 
Data Sources:
the_pile, guanaco/guanaco
Methodology:
Finetuned using DataBricks Dolly training tools
Hardware Used:
8xA100s
LLM NameQuokka 2.7B
Repository 🤗https://huggingface.co/Corianas/Quokka_2.7b 
Model Size2.7b
Required VRAM5.4 GB
Updated2026-02-06
MaintainerCorianas
Model Typegpt2
Model Files  5.4 GB   5.4 GB   0.0 GB
Supported Languagesen
Model ArchitectureGPT2LMHeadModel
Licenseapache-2.0
Transformers Version4.25.1
Tokenizer ClassGPT2Tokenizer
Beginning of Sentence Token<|endoftext|>
End of Sentence Token<|endoftext|>
Unk Token<|endoftext|>
Vocabulary Size50260
Torch Data Typebfloat16
Activation Functiongelu
Errorsreplace

Best Alternatives to Quokka 2.7B

Best Alternatives
Context / RAM
Downloads
Likes
Cerebras GPT 2.7B Alpaca SP0K / 5.4 GB86710

Rank the Quokka 2.7B 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? 52721 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