Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · Qurtana

Qwen-2.5-0.5B-MathInstruct.rev.2

Qwen-2.5-0.5B-MathInstruct.rev.2 is a 0.63B-parameter Qwen2 text-generation fine-tune from Qurtana. Its model card describes fine-tuning from unsloth/Qwen2.5-0.5B-Instruct with Unsloth and TRL.

Publisher
Qurtana
Task
text-generation
Model type
qwen2
License
apache-2.0
Library
transformers
Publication status
Accepted · not indexed

Model overview

Qwen-2.5-0.5B-MathInstruct.rev.2 is published by Qurtana as a text-generation model. The captured configuration identifies Qwen2ForCausalLM with model type qwen2, and Safetensors metadata reports 630,167,424 parameters. According to the model card, it was fine-tuned from unsloth/Qwen2.5-0.5B-Instruct.

Recorded capabilities

Compact Qwen2 fine-tune

Safetensors metadata reports about 0.63B parameters with a Qwen2ForCausalLM configuration.

Named instruction base

According to the model card, the model was fine-tuned from unsloth/Qwen2.5-0.5B-Instruct.

Unsloth and TRL training stack

The publisher states this qwen2 model was trained 2x faster with Unsloth and Hugging Face's TRL library.

Use cases in the source record

  • Small-scale conversational text-generation experiments compatible with the captured Transformers stack.
  • Fine-tune workflows building on the publisher-named unsloth/Qwen2.5-0.5B-Instruct base, as described in the model card.

Limitations and unknowns

  • No evaluation results were extracted from this record.
  • No context-window value was extracted from this record.
  • No quantization detail beyond the gguf hub tag was extracted; runtime compatibility is not specified.
  • Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: Qurtana/Qwen-2.5-0.5B-MathInstruct.rev.2

Captured: Unknown. Processed: 2026-09-07T19:34:35.740322+00:00.

Uploaded model Developed by: Qurtana License: apache-2.0 Finetuned from model : unsloth/Qwen2.5-0.5B-Instruct This qwen2 model was trained 2x faster with Unsloth and Huggingface's TRL library.

F001F002F003F004F005F006F007F008F009F010F011