Do you think koboldcpp will support Kimi-VL-A3B-Instruct and Kimi-VL-A3B-Thinking?

[SparklingDetergent]

Do you think koboldcpp will support Kimi-VL-A3B-Instruct and Kimi-VL-A3B-Thinking?

---

Kimi-VL-A3B-Instruct
https://huggingface.co/moonshotai/Kimi-VL-A3B-Instruct

Kimi-VL-A3B-Thinking
https://huggingface.co/moonshotai/Kimi-VL-A3B-Thinking

---

I was able to observe the process of gemma3 being supported in real time with llama.cpp and koboldcpp.

Therefore, we can assume that llama.cpp will first need to support creating ggufs for Kimi-VL.

I assume that koboldcpp will also support this.

I tested Kimi-VL in a hugginface demo environment and found that it is small-scale yet highly performant.

For this reason, I hope that it will eventually be available for koboldcpp as well.

However, this is just my hope.
Are you all excited about Kimi-VL?

[jabberjabberjabber]

You should discuss this on the Llamacpp page, they are the ones who have to implement it.

[SparklingDetergent]

I thought you were right too.

This theme might be useful after llama.cpp supports Kimi-VL.

I will close this for now.