TVMError: Binary was created using {relax.VMExecutable} but a loader of that name is not registered. Available loaders are relax.Executable, opencl. Perhaps you need to recompile with this runtime enabled

Questions
1

I used the MLC tool on my Mac to compile a large model, then compiled it into an Android app. When running it on my phone, I encountered the following error:

org.apache.tvm.Base$TVMError: TVMError: Binary was created using {relax.VMExecutable} but a loader of that name is not registered. Available loaders are relax.Executable, opencl. Perhaps you need to recompile with this runtime enabled.

Here are my specific steps:

  1. Install TVM Unity Compiler

python -m pip install --pre -U -f https://mlc.ai/wheels mlc-ai-nightly-cpu

  1. Install MLC LLM Python Package

python -m pip install --pre -U -f https://mlc.ai/wheels mlc-llm-nightly-cpu mlc-ai-nightly-cpu

  1. Convert Model Weights

  2. Compile Model Libraries

  3. Update mlc-llm/android/MLCChat/mlc-package-config.json

{
    "device": "android",
    "model_list": [
        {
            "model": "../../../../android/dist/deepseek-r1-1.5b-q4f16_1-MLC/",
            "model_lib": "qwen2_q4f16_1",
            "estimated_vram_bytes": 1537736704,
            "model_id": "deepseek-r1-1.5b-q4f16_1-MLC",
            "overrides": {
                "prefill_chunk_size": 768
            },
           "bundle_weight": true
        }
    ],
    "model_lib_path_for_prepare_libs": {
      "qwen2_q4f16_1": "../../../../android/dist/libs/deepseek-r1-1.5b-q4f16_1-android.tar"
    }
}

  1. Build export MLC_LLM_SOURCE_DIR=/path/to/mlc-llm mlc_llm package

  2. Build Android App

  3. Install APK

python bundle_weight.py --apk-path app/release/app-release.apk

When running on the phone, the following error occurs:

org.apache.tvm.Base$TVMError: TVMError: Binary was created using {relax.VMExecutable} but a loader of that name is not registered. Available loaders are relax.Executable, opencl. Perhaps you need to recompile with this runtime enabled.

What could be the reason?

2

please first see if you can compile model following the official instruction, the main reason of this error is because the tvm4j library is not bundled correctly Android SDK — mlc-llm 0.1.0 documentation

3

Thank you for reply. I am able to compile the model normally, and the generated files also appear to be normal.

└── lib
    └── mlc4j
        ├── build.gradle
        ├── output
        │   ├── arm64-v8a
        │   │   └── libtvm4j_runtime_packed.so
        │   └── tvm4j_core.jar
        └── src
            ├── cpp
            │   └── tvm_runtime.h
            └── main
                ├── AndroidManifest.xml
                ├── assets
                │   └── mlc-app-config.json
                └── java
                    └── ai
                        └── mlc
                            └── mlcllm
                                ├── JSONFFIEngine.java
                                ├── MLCEngine.kt
                                └── OpenAIProtocol.kt

However, the previous issue still exists.