wao/
Qwen3-30B-A3B:latest

314 5 months ago

Qwen3-30B-A3B converted to GGUF format

5 months ago

76377b58cabf · 61GB ·

model
qwen3moe
·
30.5B
·
F16
template
<|user|> {{ .Prompt }} <|assistant|>
system
You are a helpful and knowledgeable assistant.
license
Apache-2.0
params
{ "repeat_penalty": 1.1, "temperature": 0.7, "top_k": 50, "top_p": 0.9 }

Readme

Qwen3-30B-A3B (GGUF for Ollama)

This is a GGUF-formatted version of the Qwen/Qwen3-30B-A3B model, designed for compatibility with Ollama.

📌 Model Information

  • Original model: Qwen3-30B-A3B
  • Original license: Apache License 2.0
  • Converted format: GGUF
  • Converted using: transformers, llama.cpp
  • Ollama support: Yes (modelfile included)

🔧 Modifications

  • Converted the model weights to GGUF format.
  • Optional quantization may be applied (e.g., Q4_0).
  • Added Ollama-compatible modelfile.

📄 License

This project is a derived work of Qwen/Qwen3-30B-A3B, which is licensed under the Apache License 2.0.

We have not altered the model’s logic or core weights, only the format (GGUF) and packaging for Ollama compatibility.

All credit for the original model goes to the Qwen team.