Does this crate support ncnn?

[Raj2032]

Search before asking

• I have searched the Ultralytics issues and discussions and found no similar questions.

Question

I have this code:

use ultralytics_inference::{YOLOModel, InferenceConfig};

fn main()
{
    // Load your custom NCNN model (auto-detects format)
    let mut model = YOLOModel::load("custom_ncnn_model/").unwrap();
    
    // Configure exactly like Python predict params
    let mut config = InferenceConfig::default();
    config.confidence_threshold = 0.6;           // Confidence threshold
    //config. = true;          // Real-time display window
    //config.line_thickness = 2;   // Box line thickness
    config.save = false;         // No output saving
    //config.stream = true;        // Process video frame-by-frame

    // Run prediction (handles video streaming + visualization automatically)
    //model.predict_config("video.mp4", config);
    model.predict("video.mp4").unwrap();
}

However I get this error:

thread 'main' (100562) panicked at src/main.rs:6:59:
called `Result::unwrap()` on an `Err` value: ModelLoadError("Failed to load model: Load model from custom_ncnn_model/ failed:Protobuf parsing failed.")
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

Doesn't this crate support ncnn models? Do I have to use onnx instead?

Additional

No response

[UltralyticsAssistant]

👋 Hello @Raj2032, thank you for submitting a ultralytics/inference 🚀 Issue. This is an automated response to help us triage quickly, and an Ultralytics engineer will also assist soon 😊

To help us address your question efficiently, please ensure you've provided the following information:

1. For bug reports:

• A clear and concise description of the bug
• A minimum reproducible example MRE that demonstrates the issue (ideally a tiny project/repo or a minimal main.rs plus the exact model files/folder structure needed to reproduce)
• Your environment details (OS, Rust version, package versions)
• Expected behavior vs. actual behavior
• Any error messages or logs related to the issue (please include full output and, if possible, run with RUST_BACKTRACE=1)

2. For feature requests:

• A clear and concise description of the proposed feature
• The problem this feature would solve
• Any alternative solutions you've considered

3. For questions:

• Provide as much context as possible about your question
• Include any research you've already done on the topic
• Specify which parts of the documentation, if any, you've already consulted

Please make sure you've searched existing issues to avoid duplicates. If you need to add any additional information, please comment on this issue.

A couple of details that would be especially helpful here 🙏:

• The exact contents of custom_ncnn_model/ (file names + extensions) and how it was exported/created
• Your Cargo.toml dependency line(s) for ultralytics_inference (and any enabled features)
• Whether you can reproduce with a different model or with a minimal example (e.g., just YOLOModel::load(...))

Thank you for your contribution to improving our project!

[onuralpszr]

Yes, you have to use onnx for now to use rust inference, we don't have ncnn support and not the roadmap atm.

Thank you for understanding. 👍

[Raj2032]

Ok thank you mate, I have one more question, is onnx with Rust still faster than ncnn with Python on a Raspberry Pi?

…

On Monday, 19 January 2026 at 03:07, Onuralp SEZER ***@***.***> wrote: Closed [#64](#64) as not planned. — Reply to this email directly, [view it on GitHub](#64 (comment)), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/AG7P2BXRWIBVSQ23JHAXO5T4HOVVTAVCNFSM6AAAAACSCCYX3CVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMRSGEZDEOBWHAZDEMQ). You are receiving this because you were mentioned.Message ID: ***@***.***>

[onuralpszr]

Ok thank you mate, I have one more question, is onnx with Rust still faster than ncnn with Python on a Raspberry Pi?
…

I didn't test this case yet but I will do and let you know as well

cc @lakshanthad we can do small test run.

[lakshanthad]

@Raj2032 I have done tests comparing ONNX with Rust vs NCNN with Python on Raspberry Pi 5 and here are the results:

ONNX with Rust

Speed: 4.4ms preprocess, 91.8ms inference, 0.6ms postprocess per image at shape (1, 3, 384, 640)

NCNN with Python

Speed: 6.6ms preprocess, 85.2ms inference, 2.0ms postprocess per image at shape (1, 3, 640, 640)

Hope it helps :)

cc: @onuralpszr