Maxhbr/diffusion models#56
Open
maxhbr wants to merge 7 commits into
Open
Conversation
added 7 commits
June 19, 2026 14:17
Signed-off-by: Maximilian Huber <oss@maximilian-huber.de>
Add diffusiongemma-26B-A4B-it-GGUF from unsloth as an RTX model with 262k context, Q6_K quantization, and alias diffusiongemma-26B.
- Create diffusionllama-cpp overlay on thing host, building llama.cpp from PR #24423 (diffusion-gemma GPU offload support) - Add diffusionLlamaCpp option to myconfig.ai.llama-cpp module - Add diffusionCUDA device prefix support to devices.nix: - mkGuardDevice maps diffusionCUDA to "nvidia" GPU variant - backendForDevice returns "diffusion" backend tag - llamaServerForDiffusion / llamaBenchForDiffusion helpers - Wire diffusion package through scripts.nix, router.nix, llama-swap.nix via lib/default.nix parameter threading - Deploy diffusiongemma-26B-A4B-it-Q6_K on diffusionCUDA0 device DiffusionCUDA0 sets LLAMA_ARG_DEVICE=diffusionCUDA0 so the patched llama.cpp binary routes the model through the diffusion-gemma GPU offload path.
Use the real sha256 from the PR #24423 tarball instead of lib.fakeHash.
The diffusionCUDA prefix is only for Nix-side routing to pick the patched binary. The actual llama.cpp env var expects standard CUDA0.
Override postBuild to compile the llama-diffusion-cli target from PR #24423, and route diffusionCUDA devices through it instead of the standard llama-server binary.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.