github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-09 01:59:50 -05:00
Code Issues 14.7k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #9366] Magma Model Support. #6116

New Issue
Open
opened 2026-04-12 17:27:13 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-12 17:27:13 -05:00
Owner
Copy Link

Originally created by @Praveenstein on GitHub (Feb 26, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9366

Feature Request: Magma Model Support

Description:

Add support for the Magma multimodal AI agent model.

Rationale:

  • Enable local Magma execution via Ollama.
  • Simplify Magma access.

Implementation:

  • Create an Ollama modelfile for Magma.

Benefits:

  • Easy local Magma deployment.
Originally created by @Praveenstein on GitHub (Feb 26, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9366 ## Feature Request: Magma Model Support **Description:** Add support for the Magma multimodal AI agent model. **Rationale:** * Enable local Magma execution via Ollama. * Simplify Magma access. **Implementation:** * Create an Ollama modelfile for Magma. **Benefits:** * Easy local Magma deployment.
GiteaMirror added the feature request label 2026-04-12 17:27:13 -05:00
GiteaMirror commented 2026-04-12 17:27:14 -05:00
Author
Owner
Copy Link

@flywiththetide commented on GitHub (Mar 4, 2025):

Ollama is currently optimized for text-based LLMs like LLaMA, Mistral, and Phi. Magma, as a multimodal model, would require additional processing layers for handling image inputs.

Challenges of Supporting Magma in Ollama

  1. Image Processing Pipeline

    • Ollama is primarily designed for text generation, and adding image input handling would require significant backend changes.
    • Magma models often require preprocessing steps for images, which would need to be integrated into Ollama’s inference engine.

  2. Compute Requirements

    • Multimodal models often demand higher GPU memory and different optimizations compared to text-only models.
    • Efficient deployment would need support for Vision Transformers (ViTs) and embedding layers.

Potential Paths for Future Support

  • Enable Magma via an Ollama Adapter

    • If Magma is compatible with Hugging Face Transformers, it might be possible to create an adapter layer for it.

  • Use External Image Preprocessing

    • Instead of modifying Ollama’s core, a separate image preprocessing step (e.g., torchvision.transforms) could be used before passing data to a modified Ollama pipeline.

  • Community Testing & Feedback

    • If there’s enough demand, experimenting with Magma’s ONNX or TensorRT versions might be a good first step.

Would you be interested in testing a proof-of-concept Magma integration with Ollama?

<!-- gh-comment-id:2696251399 --> @flywiththetide commented on GitHub (Mar 4, 2025): Ollama is currently optimized for **text-based LLMs** like LLaMA, Mistral, and Phi. **Magma**, as a multimodal model, would require additional processing layers for handling **image inputs**. ### **Challenges of Supporting Magma in Ollama** 1. **Image Processing Pipeline** - Ollama is primarily designed for **text generation**, and adding image input handling would require **significant backend changes**. - Magma models often require **preprocessing steps** for images, which would need to be integrated into Ollama’s inference engine. 2. **Compute Requirements** - Multimodal models often demand **higher GPU memory** and **different optimizations** compared to text-only models. - Efficient deployment would need support for **Vision Transformers (ViTs) and embedding layers**. ### **Potential Paths for Future Support** - **Enable Magma via an Ollama Adapter** - If Magma is compatible with **Hugging Face Transformers**, it might be possible to create an **adapter layer** for it. - **Use External Image Preprocessing** - Instead of modifying Ollama’s core, a **separate image preprocessing step** (e.g., `torchvision.transforms`) could be used before passing data to a modified Ollama pipeline. - **Community Testing & Feedback** - If there’s enough demand, experimenting with Magma’s **ONNX or TensorRT** versions might be a good first step. Would you be interested in testing a **proof-of-concept Magma integration** with Ollama?
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label feature request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#6116
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?