Modelpublic

SoftDream11/Qwen3.6-Solidity-27B-GGUF

sourceHugging Faceagpl-3.0updated 5mo agoView on Hugging Face

0likes65downloads

Model Card

Qwen3.6-Solidity-27B-GGUF

This repository contains GGUF quantized versions of samscrack/Qwen3.6-Solidity-27B, a model specifically fine-tuned for Solidity development, smart contract auditing, and blockchain security.

Available Files

I have provided multiple quantization levels to balance performance and resource requirements:

Filename	Method	Bits	Size	Description
`Qwen3.6-Solidity-27B-Q6_K.gguf`	Q6_K	6	~22.5 GB	Balanced: Extremely low loss, near-identical to FP16.
`Qwen3.6-Solidity-27Bgguf`	full	16	~51.0 G	Balanced: Full quality.

Usage

These files are compatible with any client supporting llama.cpp, including:

—LM Studio
—Ollama
—Text-Generation-WebUI
—LocalAI

Prompt Template (Qwen)

This model follows the ChatML format. For best results, use:

text

<|im_start|>system
You are a senior Solidity security researcher and expert developer.<|im_end|>
<|im_start|>user
{your_prompt}<|im_end|>
<|im_start|>assistant