LLM EXPLORER 59,962 MODELS INDEXED

Ct2fast Codegen2 1B by michaelfeil

By michaelfeil · 7 downloads
HF Repo ↗ Copy URL

Ct2fast Codegen2 1B is an open-source language model by michaelfeil. Features: 1b LLM, VRAM: 2GB, License: apache-2.0, LLM Explorer Score: 0.07.

  Arxiv:2305.02309   Ctranslate2   Endpoints compatible   Float16   Int8   Region:us
Model Card on HF 🤗: https://huggingface.co/michaelfeil/ct2fast-codegen2-1B 

Ct2fast Codegen2 1B Parameters and Internals

Model Type 
autoregressive, program synthesis, causal language model
Use Cases 
Areas:
code completion, program synthesis
Applications:
commercial applications, research
Primary Use Cases:
text generation for programming languages, program synthesis
Limitations:
Works best with English prompts and may have biased outputs due to the training dataset.
Additional Notes 
This quantized version maintains the original licensing conditions from the Salesforce repo.
Supported Languages 
c (unknown), c++ (unknown), c-sharp (unknown), dart (unknown), go (unknown), java (unknown), javascript (unknown), kotlin (unknown), lua (unknown), php (unknown), python (unknown), ruby (unknown), rust (unknown), scala (unknown), shell (unknown), sql (unknown), swift (unknown), typescript (unknown), vue (unknown)
Training Details 
Data Sources:
deduplicated version of the Stack dataset (v1.1)
Data Volume:
unknown
Methodology:
cross-entropy loss for causal language modeling and file-level span corruption.
Model Architecture:
Unknown
Input Output 
Input Format:
Tokenized input format necessary
Accepted Modalities:
text
Output Format:
Tokenized text
Performance Tips:
Quantization and appropriate hardware usage can significantly boost performance.
Release Notes 
Version:
unknown
Date:
2023-05-21
Notes:
Converted using ct2-transformers-converter with quantization float16.
LLM NameCt2fast Codegen2 1B
Repository 🤗https://huggingface.co/michaelfeil/ct2fast-codegen2-1B 
Model Size1b
Required VRAM2 GB
Updated2026-06-29
Maintainermichaelfeil
Model Files  2.0 GB
Model ArchitectureAutoModel
Licenseapache-2.0
Model Max Length1024
Tokenizer ClassGPT2Tokenizer

Best Alternatives to Ct2fast Codegen2 1B

Best Alternatives
Context / RAM
Downloads
Likes
Waiter 1B Lora0K / 0.1 GB271
Prem 1B Chat ORPO0K / 0.3 GB01
Prem 1B ORPO0K / 0.6 GB01
Medical0K / 0 GB150
MiniMaid L10K / 0 GB81
Llama 3.2 1B Creative Lora0K / 0 GB04
Triangulum 1B GGUF0K / 0.8 GB431
Llama Deepsync 1B GGUF0K / 0.8 GB103
... SmolTalk 3.2 1B Instruct GGUF0K / 0.8 GB1074
Mergekit Slerp Ozbsicw GGUF0K / 2.5 GB6260
Note: green Score (e.g. "73.2") means that the model is better than michaelfeil/ct2fast-codegen2-1B.