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

46 lines
1.2 KiB
JSON

{
"title": "Runpod Worker Ollama",
"description": "A serverless Ollama Worker for Runpod",
"type": "serverless",
"category": "language",
"iconUrl": "https://ollama.com/public/ollama.png",
"config": {
"runsOn": "GPU",
"gpuCount": 1,
"gpuIds": "AMPERE_16,AMPERE_24,ADA_24",
"containerDiskInGb": 20,
"presets": [],
"env": [
{
"key": "OLLAMA_MODEL_NAME",
"input": {
"name": "Model Name",
"type": "string",
"description": "Name of a model to preload",
"default": "phi3",
"advanced": false
}
},
{
"key": "MAX_CONCURRENCY",
"input": {
"name": "Max Concurrency",
"type": "number",
"description": "Maximum number of concurrent requests to handle (default: 8)",
"default": 8,
"advanced": true
}
},
{
"key": "OLLAMA_NUM_PARALLEL",
"input": {
"name": "Parallel Requests",
"type": "string",
"description": "Maximum number of concurrent requests to handle (default: 4 or 1)",
"default": "",
"advanced": true
}
}
]
}
}