diffusion models!
please update llama.cpp first:
https://github.com/ggml-org/llama.cpp/pull/14644
verified models in the pull request: Dream 7B instruct and DiffuCoder-7B-cpGRPO
but maybe you could try ;)
https://huggingface.co/apple/DiffuCoder-7B-Base
https://huggingface.co/apple/DiffuCoder-7B-Instruct
https://huggingface.co/apple/DiffuCoder-7B-cpGRPO
https://huggingface.co/Dream-org/DreamOn-v0-7B
https://huggingface.co/Dream-org/Dream-v0-Base-7B
https://huggingface.co/Dream-org/Dream-v0-Instruct-7B
https://huggingface.co/Dream-org/Dream-Coder-v0-Base-7B
https://huggingface.co/Dream-org/Dream-Coder-v0-Instruct-7B
@mradermacher Please update to latest llama.cpp of ouer fork and queue them. I'm closely followed this PR as well and am supper excited about diffusion based LLMs. I especially love how nicely you can visualize the entire diffusion process.
I queued them all now but as worker +PLEASE-UPDATE-LLAMACPP. You see them on http://hf.tst.eu/status.html now but they will be stuck in the wait queue untill @mradermacher finds time to update llama.cpp.