Files
inference-worker/.runpod/tests.json
T

24 lines
538 B
JSON

{
"tests": [
{
"name": "execute_a_prompt",
"input": {
"prompt": "Hi! Who are you?"
},
"timeout": 120000
}
],
"config": {
"gpuTypeId": "NVIDIA GeForce RTX 4090",
"gpuCount": 1,
"env": [
{
"key": "LLAMA_SERVER_CMD_ARGS",
"value": "-hf unsloth/gemma-3-270m-it-GGUF:Q6_K --ctx-size 4096"
}
],
"allowedCudaVersions": [
"12.8"
]
}
}