{"data":{"slug":"defilantech-llmkube","name":"LLMKube","tagline":"Kubernetes operator for self-hosted LLM inference across heterogeneous GPU fleets","github_url":"https://github.com/defilantech/LLMKube","owner":"defilantech","repo":"LLMKube","owner_avatar_url":"https://avatars.githubusercontent.com/u/69058620?v=4","primary_language":"Go","stars":210,"forks":33,"topics":["ai","apple-silicon","autoscaling","edge-computing","gguf","gpu","homelab","inference","kubernetes","kubernetes-operator","llama-cpp","llm","local-llm","metal","mlx","multi-gpu","nvidia","self-hosted","tgi","vllm"],"archived":false,"github_pushed_at":"2026-09-17T14:44:50+00:00","maintenance_label":"Very active","stars_delta_30d":27,"url":"https://www.graphcanon.com/tools/defilantech-llmkube","markdown_url":"https://www.graphcanon.com/tools/defilantech-llmkube.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/defilantech-llmkube","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=defilantech-llmkube","description":"Kubernetes operator for self-hosted LLM inference across a heterogeneous GPU fleet: NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal. Runtimes: llama.cpp, vLLM, TGI, mlx-server. Multi-GPU sharding, model caching, OpenAI-compatible endpoints. Apache-2.0, run across homelab and on-prem fleets, actively developed.","homepage_url":"https://llmkube.com","license":"Apache-2.0","open_issues":98,"watchers":3,"ai_summary":"LLMKube is a Kubernetes operator designed for deploying and managing self-hosted large language model inference services on a variety of GPU architectures, including NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal. It supports multiple runtimes like llama.cpp, vLLM, TGI, and mlx-server, and offers features such as multi-GPU sharding, model caching, and OpenAI-compatible endpoints.","readme_excerpt":"# Install the CLI\nbrew install defilantech/tap/llmkube\n\n---\n\n# Install the operator on any K8s cluster\nhelm repo add llmkube https://defilantech.github.io/LLMKube\nhelm install llmkube llmkube/llmkube --namespace llmkube-system --create-namespace\n\n---\n\n## License\n\nApache 2.0 — see [LICENSE](LICENSE).","github_created_at":"2025-11-12T22:53:23+00:00","created_at":"2026-07-11T23:11:10.734138+00:00","updated_at":"2026-09-20T04:29:41.109302+00:00","categories":[{"slug":"inference-serving","name":"Inference & Serving","url":"https://www.graphcanon.com/categories/inference-serving","markdown_url":"https://www.graphcanon.com/categories/inference-serving.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/inference-serving"}],"tags":[{"slug":"ai","name":"ai"},{"slug":"apple-silicon","name":"apple-silicon"},{"slug":"autoscaling","name":"autoscaling"},{"slug":"edge-computing","name":"edge-computing"},{"slug":"gguf","name":"gguf"},{"slug":"gpu","name":"gpu"},{"slug":"homelab","name":"homelab"},{"slug":"inference","name":"inference"}],"trust":{"provenance":{"is_fork":false,"github_id":1095330682,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-18T08:17:44.982Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":30,"days_since_push":0,"last_release_at":"2026-09-17T08:21:26Z","stars_delta_30d":27,"open_issues_delta_30d":21},"security_summary":{"status":"findings","scanner":"osv@v1","low_count":3,"high_count":0,"last_scan_at":"2026-09-18T08:17:46.633Z","medium_count":0,"scan_profile":"deps","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-18T08:17:43.894Z"},"deploy":{"source":"dockerfile:Dockerfile","self_host":true,"observed_at":"2026-09-18T08:17:43.894Z","managed_saas":false},"languages":{"value":["go"],"source":"github.language","observed_at":"2026-09-18T08:17:43.894Z"},"has_docker":{"value":true,"source":"dockerfile:Dockerfile","observed_at":"2026-09-18T08:17:43.894Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-09-18T08:17:43.894Z"}},"decision_facts":{"hosting":null,"pricing":{"model":"freemium","summary":"Free and open-source under the Apache-2.0 license."},"requirements":{"notes":["Requires a Kubernetes cluster for installation and operation.","Supports deployment on a variety of GPU architectures, including NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal."],"min_ram_gb":null,"requires_docker":false},"constraints":{"min_ram_gb":null,"pricing_model":"freemium","requires_docker":false},"when_to_use":["When you need to deploy and manage self-hosted large language model inference services across a variety of GPU architectures, including NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal.","If you require support for multiple runtimes like llama.cpp, vLLM, TGI, and mlx-server, and features such as multi-GPU sharding, model caching, and OpenAI-compatible endpoints.","For environments that need to run across homelab and on-prem fleets, LLMKube is actively developed and can be a good fit."],"when_not_to_use":["If your infrastructure does not include a Kubernetes cluster, as LLMKube is designed to operate within a Kubernetes environment.","If you are looking for a solution that does not support a heterogeneous mix of GPU architectures, LLMKube is specifically designed to work with NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal.","If you do not require or prefer not to use the specific runtimes supported by LLMKube, such as llama.cpp, vLLM, TGI, and mlx-server."],"source":"enrich:decision_facts","observed_at":"2026-09-18T09:56:38.486Z"},"constraint_facets":{"min_ram_gb":null,"pricing_model":"freemium","requires_docker":false},"decision_summary":[{"label":"Pricing","value":"freemium - Free and open-source under the Apache-2.0 license."},{"label":"Requirements","value":"Requires a Kubernetes cluster for installation and operation.; Supports deployment on a variety of GPU architectures, including NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal."},{"label":"Adopt for","value":"LLMKube is a Kubernetes operator for deploying and managing self-hosted large language model inference services on various GPU architectures, including NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal. It supports runtim"},{"label":"License detail","value":"Apache-2.0"}]}}