-
Notifications
You must be signed in to change notification settings - Fork 932
Description
我想我的平板cpu是支持android demo运行的,所以我就尝试,但崩溃了。我也尝试了 plugin_id = "cpu_gpu", plugin_id = "cpu"都是这个问题。
我的平板信息:
设备基本信息
- 设备型号 : DMG-W00
- 制造商 : HUAWEI
- 芯片平台 : Kirin9000WM
CPU 信息
- 架构 : ARMv8 (64位)
- 核心配置 : 6核心
- 3个 Cortex-A710 (大核,CPU implementer: 0x41, CPU part: 0xd46)
- 3个 Cortex-A510 (小核,CPU implementer: 0x48, CPU part: 0xd42/0xd02)
- 特性 : 支持加密指令集、向量计算等高级特性
GPU 信息
- GPU 型号 : Maleoon 910 (华为自研 GPU)
- OpenGL ES 版本 : 3.2
- 驱动版本 : B236.95f3eca3c57559aa9ed21461acd7e5da485113ec
NPU 信息
- NPU 型号 : 华为达芬奇 NPU (Da Vinci NPU)
- HIAI 版本 : 100.629.020.110
- 服务状态 : hiaiserver 运行中
我的demo代码:
val modelPath = "/data/data/ai.codia.x.java.nexaaiproject/files/models/OmniNeural-4B-mobile"
lifecycleScope.launch {
VlmWrapper.builder()
.vlmCreateInput(
VlmCreateInput(
model_name = "omni-neural",
model_path = modelPath,
config = ModelConfig(
max_tokens = 2048,
enable_thinking = false
),
plugin_id = "cpu_gpu"
)
)
.build()
.onSuccess { vlmWrapper = it }
.onFailure { error ->
println("Failed to load model: ${error.message}")
}
}
我的崩溃log
[STDERR] === NEXA SDK: stderr redirection test - this should appear in logcat ===
dlopen failed: library "/data/app/~~sdypg3dPCZkqadWsueWqvg==/ai.codia.x.java.nexaaiproject-9t_41Et6Ph8f6Dhobi6S2g==/lib/arm64/libnexa_plugin_cpu_gpu.so" not found
java.lang.RuntimeException: Unable to start activity ComponentInfo{ai.codia.x.java.nexaaiproject/ai.codia.x.java.nexaaiproject.MainActivity}: java.lang.RuntimeException: Cannot find cpu_gpu in /data/app/~~sdypg3dPCZkqadWsueWqvg==/ai.codia.x.java.nexaaiproject-9t_41Et6Ph8f6Dhobi6S2g==/lib/arm64
