Our current primary focus centers exclusively on creating quantizations for Qwen and DeepSeek. Alongside these efforts, a segment of our product team is actively pursuing fine-tuning developments.

The response to our recent work has been incredible, as the Qtum repository on huggingface has accumulated more than 20,000 downloads in fewer than 10 days. The DeepSeekV4Flash 0731 GGUF model has been a major highlight, securing 18,000 of those downloads on its own.

You can explore the complete download statistics by visiting https://huggingface.co/qtum/models?sort=downloads