Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · QuyXuan

documents-master-3B

documents-master-3B is a 3.21B-parameter Llama-family text-generation fine-tune from QuyXuan, finetuned from unsloth/Llama-3.2-3B-Instruct with Unsloth and TRL.

Publisher
QuyXuan
Task
text-generation
Model type
llama
License
apache-2.0
Library
transformers
Publication status
Accepted · not indexed

Model overview

documents-master-3B is published by QuyXuan as a Llama-based text-generation fine-tune. The captured configuration identifies LlamaForCausalLM and Safetensors metadata reports 3,212,749,824 parameters. According to the model card, it is finetuned from unsloth/Llama-3.2-3B-Instruct using Unsloth and Hugging Face's TRL library, and card data records apache-2.0.

Recorded capabilities

Llama 3.2 3B-class fine-tune

According to the model card, the model is finetuned from unsloth/Llama-3.2-3B-Instruct, with captured metadata reporting about 3.21B parameters on a Llama architecture.

Unsloth and TRL workflow

According to the model card, training used Unsloth with Hugging Face's TRL library, described by the publisher as 2x faster.

Transformers conversational setup

Hub data records Transformers support with conversational and text-generation-inference tags and endpoints compatibility.

Use cases in the source record

  • Conversational text-generation work building on the documented Llama-3.2-3B-Instruct finetune lineage.
  • SFT-style experiments using the publisher's documented Unsloth and TRL workflow.

Limitations and unknowns

  • No evaluation results were extracted from this record.
  • No context-window value was extracted from this record.
  • Provider state is historical snapshot data and should be refreshed before being presented as current.

Source and provenance

Source: QuyXuan/documents-master-3B

Captured: Unknown. Processed: 2026-09-07T19:35:51.211448+00:00.

Uploaded model Developed by: jCool10 License: apache-2.0 Finetuned from model : unsloth/Llama-3.2-3B-Instruct This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

F001F002F003F004F005F006F007F008F009F010F011