Wire the trained hardware-recommender model into model recommendations #47
voidstackloopAnnotations
3 errors and 1 warning
|
test
Process completed with exit code 1.
|
|
src/system-specs.test.ts > recommendModels > sums VRAM across multiple GPUs instead of using only the first one:
app/src/system-specs.test.ts#L70
AssertionError: expected 42.2 to be close to 42.24, received difference is 0.03999999999999915, but expected 0.005
❯ src/system-specs.test.ts:70:37
|
|
src/local-server-manager.test.ts > describeSpawnFailure > points vLLM failures at installation rather than endpoint configuration:
app/src/local-server-manager.test.ts#L86
AssertionError: expected 'Couldn\'t launch vLLM. Install vLLM i…' to match /pip install vllm/
- Expected:
/pip install vllm/
+ Received:
"Couldn't launch vLLM. Install vLLM in Linux/WSL and verify that its CUDA or ROCm requirements match the GPU driver."
❯ src/local-server-manager.test.ts:86:46
|
|
test
Node.js 20 is deprecated. The following actions target Node.js 20 but are being forced to run on Node.js 24: actions/checkout@v4, actions/setup-node@v4. For more information see: https://github.blog/changelog/2025-09-19-deprecation-of-node-20-on-github-actions-runners/
|