#Any plans for Unsloth Dynamic 2.0 GGUFs for DeepSeek V4 Flash?

2 messages · Page 1 of 1 (latest)

feral wind
#

Hi team — first off, thanks for the incredible work on Dynamic 2.0 quantization. The results on DeepSeek-R1 and V3-0324 are fantastic.

Now that DeepSeek V4 Flash is out (and you've already hosted the raw model on HF), I'm wondering: do you have plans to release Dynamic 2.0 GGUFs for DeepSeek V4 Flash? And if so, any rough ETA?

A lot of us are eager to run this locally with your quantization — the accuracy-per-bit tradeoff with Dynamic 2.0 would be a huge win for self-hosted inference.

Thanks!

frank musk
#

Hello we werre originally working on the PR but it's more complicated than originally antitipcated