LLM EXPLORER 61,836 MODELS INDEXED

Ct2fast WizardCoder 15B V1.0 by michaelfeil

By michaelfeil · 6 downloads
HF Repo ↗ Copy URL

Ct2fast WizardCoder 15B V1.0 is an open-source language model by michaelfeil. Features: 15b LLM, VRAM: 15.6GB, License: bigcode-openrail-m, LLM Explorer Score: 0.07.

  Ctranslate2   Endpoints compatible   Float16   Int8   Region:us
Model Card on HF 🤗: https://huggingface.co/michaelfeil/ct2fast-WizardCoder-15B-V1.0 

Ct2fast WizardCoder 15B V1.0 Parameters and Internals

Model Type 
code generation
Use Cases 
Areas:
research
Applications:
code-related tasks
Primary Use Cases:
code generation
Limitations:
Cannot guarantee output accuracy; not for commercial use
Considerations:
Usage influenced by instructive coding tasks
Additional Notes 
Quantized version reduces memory by 2x-4x, supports int8/float16 on CUDA or CPU.
Training Details 
Data Sources:
evolved code instructions
Data Volume:
78k instructions
Methodology:
Evol-Instruct method, fine-tuning with modified code
Context Length:
2048
Model Architecture:
Modified Transformer
Input Output 
Input Format:
code-related instruction prompts
Accepted Modalities:
text
Output Format:
generated code text
Release Notes 
Version:
v1.0
Date:
2023-06-15
Notes:
First quantized version
LLM NameCt2fast WizardCoder 15B V1.0
Repository 🤗https://huggingface.co/michaelfeil/ct2fast-WizardCoder-15B-V1.0 
Model Size15b
Required VRAM15.6 GB
Updated2026-07-13
Maintainermichaelfeil
Model Files  15.6 GB
Model ArchitectureAutoModel
Licensebigcode-openrail-m
Model Max Length2048
Tokenizer ClassGPT2Tokenizer
Vocabulary Size49152

Best Alternatives to Ct2fast WizardCoder 15B V1.0

Best Alternatives
Context / RAM
Downloads
Likes
WizardCoder 15B 1.0 GGML0K / 10.7 GB11115
Note: green Score (e.g. "73.2") means that the model is better than michaelfeil/ct2fast-WizardCoder-15B-V1.0.