9877ac1c70
YAML file describing available AI models together with a new AI_MODEL option for simple retrieval of the available models.
4 lines
988 B
YAML
4 lines
988 B
YAML
- { id: 8b0b55b3-fb99-415f-8441-3abef2c2a200, model_name: test-model, enabled: true, locations: [hetzner-fsn1], vm_size: standard-gpu-6 , storage_volumes: [{encrypted: true, size_gib: 80}, {read_only: true, image: ai-model-test-model}], boot_image: ai-ubuntu-2404-nvidia, engine: vllm, engine_params: "" }
|
|
- { id: 04ba0d97-859b-46ba-a90b-36a7c7900d4b, model_name: gemma-2-2b-it, enabled: true, locations: [hetzner-fsn1], vm_size: standard-gpu-6 , storage_volumes: [{encrypted: true, size_gib: 80}, {read_only: true, image: ai-model-gemma-2-2b-it}], boot_image: ai-ubuntu-2404-nvidia, engine: vllm, engine_params: "" }
|
|
- { id: acc50340-c036-44ff-85a2-c5b7c8823e2a, model_name: llama-3-2-3b-it, enabled: true, locations: [hetzner-fsn1], vm_size: standard-gpu-6 , storage_volumes: [{encrypted: true, size_gib: 80}, {read_only: true, image: ai-model-llama-3-2-3b-it}], boot_image: ai-ubuntu-2404-nvidia, engine: vllm, engine_params: "--gpu-memory-utilization 0.95 --max-model-len 90000" }
|