Viper is a custom language model based on Qwen2.5-7B-Instruct, fine-tuned by viperprojects47 for custom software development assistance. It is distributed as a 4-bit quantized GGUF for use with llama.cpp, Ollama, and any OpenAI-compatible server.

Files

File	Size	Quant	Description
`viper.gguf`	4.4GB	Q4_K_M	Quantized GGUF for llama.cpp etc.

Run with llama.cpp

llama-server -m viper.gguf --port 8080 -c 2048 -ngl 99

Then call the OpenAI-compatible endpoint:

bash

curl http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"messages": [{"role": "user", "content": "Who developed you?"}]}'

Run with Ollama

ollama run hf.co/USERNAME/viper-ai

or create a local model:

ollama create viper -f ./Modelfile

Model details

Base architecture: Qwen2.5-7B-Instruct
Format: GGUF (Quantized) — Q4_K_M
Fine-tuned by: viperprojects47
System prompt: "You are Viper, a personal AI coding model fine-tuned for custom software development by viperprojects47."

viperprojects47/viper-ai

Viper AI — 7B

Files

Run with llama.cpp

Run with Ollama

Model details

viperprojects47/
viper-ai