ODS V3 Pre-Release: Public testing and refinement ahead of the official V3 launch. Turn your PC, Mac, or Linux box into a private AI server.
Python ai-agentsamdcomfyuidockerllama-cppllmlocal-ain8nnvidiaopen-webuiragself-hostedspeech-to-textstrix-halotext-to-speechworkflow-automation
{
"id": "nvidia",
"llm_engine": "llama-server",
"service_name": "llama-server",
"public_api_port": 8080,
"public_health_url": "http://127.0.0.1:8080/health",
"provider_name": "local-llama",
"provider_url": "http://llama-server:8080/v1",
"runtime": {
"llama_server": {
"container_images_by_compute": {
"12.1": "ghcr.io/ggml-org/llama.cpp:server-cuda13-b11429@sha256:9d6ca9a08b5e0c0a6d2560bff7d4c76bdf946776bbde6a20d9d6403594d93a9f"
},
"windows_arm64": {
"release_tag": "b11429",
"flavor": "win-cuda-13.4-arm64",
"backend": "cuda",
"policy_key": "windows-native-cuda",
"archives": [
{
"asset": "llama-b11429-bin-win-cuda-13.4-arm64.zip",
"sha256": "930cff325a0eb15e1232d5b5275b818c76c9c32a398788f18bb05273d3110c72",
"size": 145270624
},
{
"asset": "cudart-llama-bin-win-cuda-13.4-arm64.zip",
"sha256": "642dcde8805b3e3165ca710a5443b3b4044b27d96bd3ee3132473988c9bcb774",
"size": 153262407
}
],
"required_files": [
"llama-server.exe",
"llama-server-impl.dll",
"llama-common.dll",
"llama.dll",
"mtmd.dll",
"ggml.dll",
"ggml-base.dll",
"ggml-cpu.dll",
"ggml-cuda.dll",
"libomp.dll",
"cudart64_13.dll",
"cublas64_13.dll",
"cublasLt64_13.dll"
],
"min_driver_version": "616.41",
"cuda_archs_by_compute": {
"12.1": ["1210", "121", "121a"]
},
"model_ladder": ["qwen3.6-35b-a3b-ud-q4"],
"fallback_models": ["qwen3.5-2b-q4"]
}
}
}
}