Skip to content

[BUG]: have not yet support gemma3? #1129

New issue
New issue
Closed
Closed
[BUG]: have not yet support gemma3?#1129

Description

@nipeone
nipeone
opened on Mar 18, 2025

Description

llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'gemma3'

Reproduction Steps

i use model unsloth/gemma-3-4b-it-GGUF

Environment & Configuration

  • Operating system: windows 11
  • .NET runtime version: .net 8.0
  • LLamaSharp version: 0.21.0
  • CUDA version (if you are using cuda backend):
  • CPU & GPU device: CPU

Known Workarounds

No response

Activity

  1. martindevans commented on Mar 18, 2025

    @martindevans
    martindevans
    on Mar 18, 2025
    Member

    There's already a PR open to update the llama.cpp version, once that's merged that should support gemma3 :)

  2. martindevans commented on Mar 20, 2025

    @martindevans
    martindevans
    on Mar 20, 2025
    Member

    The latest version on nuget should support Gemma3 now.

  3. nipeone
    closed this as completedon Mar 21, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions