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

GPT Bigcode Santacoder by bigcode

 »  All LLMs  »  bigcode  »  GPT Bigcode Santacoder   URL Share it on

  48319   
  Autotrain compatible   Code   Codegen   Dataset:bigcode/the-stack   Gpt bigcode   Model-index   Pytorch   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/bigcode/gpt_bigcode-santacoder 

GPT Bigcode Santacoder Benchmarks

ARC: 21.16 vs 96.7 (so35)-78.1%
HellaSwag: 30.84 vs 95.3 (gpt4)-67.6%
MMLU: 24.97 vs 88.3 (so35)-71.7%
TruthfulQA: 45.64 vs 59 (gpt4)-22.6%
WinoGrande: 47.83 vs 87.5 (gpt4)-45.3%
GSM8K: 0.53 vs 96.4 (so35)-99.5%
LLME Score: 0.18197
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
GPT Bigcode Santacoder (bigcode/gpt_bigcode-santacoder)
🌟 Advertise your project 🚀

GPT Bigcode Santacoder Parameters and Internals

Model Type 
text-generation
Use Cases 
Areas:
Source code generation
Applications:
Software development, Code completion
Primary Use Cases:
Generating code snippets with provided context
Limitations:
Cannot guarantee working code, May produce inefficient or bug-prone code
Considerations:
Model is trained on open-source code, ensure adherence to licensing.
Additional Notes 
Model pretraining included filtering for permissive licenses and can generate code verbatim, requiring license compliance.
Supported Languages 
Python (High), Java (High), JavaScript (High)
Training Details 
Data Sources:
GitHub code filtered for permissive licenses
Data Volume:
236 billion tokens
Methodology:
Filled for permissive licenses; uses multi-query attention and Fill-in-the-Middle objective
Training Time:
6.2 days
Hardware Used:
96 Tesla V100 GPUs
Model Architecture:
GPT-2 model with multi-query attention
Input Output 
Input Format:
Model expects code-like inputs along with comments or function signatures.
Accepted Modalities:
text
Output Format:
Code completions
Performance Tips:
Ensure inputs are appropriately structured to resemble typical source code prompts.
LLM NameGPT Bigcode Santacoder
Repository 🤗https://huggingface.co/bigcode/gpt_bigcode-santacoder 
Model Size1.1b
Required VRAM2.2 GB
Updated2025-09-23
Maintainerbigcode
Model Typegpt_bigcode
Model Files  2.2 GB   2.2 GB
Supported Languagescode
Generates CodeYes
Model ArchitectureGPTBigCodeForCausalLM
Licenseopenrail
Model Max Length2048
Transformers Version4.28.0.dev0
Tokenizer ClassGPT2TokenizerFast
Vocabulary Size49280
Activation Functiongelu_pytorch_tanh
Errorsreplace

Rank the GPT Bigcode Santacoder 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? 51534 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