Nsql 350M by NumbersStation

 ยป  All LLMs  ยป  NumbersStation  ยป  Nsql 350M   URL Share it on

  Autotrain compatible   Codegen   Endpoints compatible   Pytorch   Region:us

Nsql 350M Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Nsql 350M (NumbersStation/nsql-350M)

Nsql 350M Parameters and Internals

Model Type 
autoregressive, SQL generation
Use Cases 
Primary Use Cases:
text-to-SQL generation tasks
Limitations:
Works best with prompt format defined for SELECT queries
Additional Notes 
The model works best with the prompt format defined below and outputting SELECT queries.
Training Details 
Data Sources:
The Stack, more than 20 public sources across the web
Data Volume:
1M training samples for general SQL queries
Methodology:
Pre-trained on a dataset of general SQL queries and fine-tuned on text-to-SQL pairs
Hardware Used:
80GB A100s
LLM NameNsql 350M
Repository ๐Ÿค—https://huggingface.co/NumbersStation/nsql-350M 
Model Size350m
Required VRAM1.5 GB
Updated2025-02-22
MaintainerNumbersStation
Model Typecodegen
Model Files  1.5 GB
Generates CodeYes
Model ArchitectureCodeGenForCausalLM
Licensebsd-3-clause
Model Max Length2048
Transformers Version4.28.1
Tokenizer ClassGPT2Tokenizer
Vocabulary Size51200
Torch Data Typefloat32
Activation Functiongelu_new

Best Alternatives to Nsql 350M

Best Alternatives
Context / RAM
Downloads
Likes
Ft Calculator0K / 0.6 GB820
...no Finetuned Python 18K Alpaca0K / 0.8 GB1091
...o Python 18K Alpaca Runpod.net0K / 0.7 GB91
...hs Finetuned Python 18K Alpaca0K / 0.8 GB90
... Powershell Codegen 350M Multi0K / 1.4 GB1081
Codegen 350M Mono Java Merged0K / 1.5 GB1090
...en 350M Mono 18K Alpaca Python0K / 0.7 GB112
...om Functions Dataset Python V20K / 1.5 GB20361
Codegen 350M Mono0K /  GB3096
DailyChat 350M0K / 1.5 GB1180
Note: green Score (e.g. "73.2") means that the model is better than NumbersStation/nsql-350M.

Rank the Nsql 350M Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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? 43470 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227