{ "archive_path": "archive/1790892380.785368", "base_url": "github.com/Vibra-Ingenn/Janus", "basename": "Janus", "bookmarked_date": "2026-10-01 22:06", "canonical": { "archive_org_path": "https://web.archive.org/web/github.com/Vibra-Ingenn/Janus", "dom_path": "output.html", "favicon_path": "favicon.ico", "git_path": "git/", "google_favicon_path": "https://www.google.com/s2/favicons?domain=github.com", "headers_path": "headers.json", "htmltotext_path": "htmltotext.txt", "index_path": "index.html", "media_path": "media/", "mercury_path": "mercury/content.html", "pdf_path": "output.pdf", "readability_path": "readability/content.html", "screenshot_path": "screenshot.png", "singlefile_path": "singlefile.html", "warc_path": "warc/", "wget_path": "github.com/Vibra-Ingenn/Janus.html" }, "domain": "github.com", "extension": "", "hash": "K7TPWXPZ2ZS07GV2PZRX", "history": { "archive_org": [], "dom": [], "favicon": [ { "cmd": [ "curl", "--silent", "--location", "--compressed", "--max-time", "120", "--output", "favicon.ico", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://www.google.com/s2/favicons?domain=github.com" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-10-01T22:07:19.658409+00:00", "index_texts": null, "output": "favicon.ico", "pwd": "/data/archive/1790892380.785368", "schema": "ArchiveResult", "start_ts": "2026-10-01T22:07:19.056570+00:00", "status": "succeeded" } ], "git": [], "headers": [ { "cmd": [ "curl", "--silent", "--location", "--compressed", "--head", "--max-time", "120", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://github.com/Vibra-Ingenn/Janus" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-10-01T22:07:20.059176+00:00", "index_texts": null, "output": "headers.json", "pwd": "/data/archive/1790892380.785368", "schema": "ArchiveResult", "start_ts": "2026-10-01T22:07:19.667300+00:00", "status": "succeeded" } ], "htmltotext": [], "media": [], "mercury": [ { "cmd": [ "/app/node_modules/@postlight/parser/cli.js", "https://github.com/Vibra-Ingenn/Janus" ], "cmd_version": "1.0.0", "end_ts": "2026-10-01T22:07:35.941350+00:00", "index_texts": null, "output": "mercury", "pwd": "/data/archive/1790892380.785368", "schema": "ArchiveResult", "start_ts": "2026-10-01T22:07:30.390998+00:00", "status": "succeeded" } ], "pdf": [], "readability": [ { "cmd": [ "/app/node_modules/readability-extractor/readability-extractor", "/tmp/tmps77ewoox", "https://github.com/Vibra-Ingenn/Janus" ], "cmd_version": "0.0.11", "end_ts": "2026-10-01T22:07:30.381691+00:00", "index_texts": [ "Janus is a single Go binary that runs .gguf models on your machine (GPU or CPU) and exposes an OpenAI-compatible API. No Python, no Docker, no Ollama required.\nUse it your way: call it from the command line (curl, PowerShell, scripts), wire it into Cursor / Cline / any OpenAI client \u2014 same local models, whatever workflow fits you.\n\nWhat you get\n\nLocal inference \u2014 llama.cpp via Vulkan (AMD / Intel / NVIDIA) or CPU fallback\nOpenAI-compatible API \u2014 /v1/chat/completions, /v1/models\nHot-swap models \u2014 change .gguf without restarting\nThinking model support \u2014 reasoning split into reasoning_content\nChat template auto-detection \u2014 uses the template from GGUF metadata\nZero dependencies \u2014 one .exe on Windows, no Python, no Docker\n\n\nRequirements\n\n\n\nPlatform\nWhat you need\n\n\n\n\nWindows (primary)\nWindows 10/11, Go 1.22+, Vulkan-capable GPU recommended\n\n\nLinux\nGo 1.22+, Vulkan or CPU\n\n\nmacOS\nGo 1.22+, CPU backend (Vulkan varies by hardware)\n\n\n\nDisk: plan for the model size (often 2\u20138 GB per model) plus ~50 MB for Janus + llama.dll.\n\nQuick start (Windows)\n1. Clone and build\ngit clone https://github.com/Vibra-Ingenn/Janus.git\ncd Janus\n.\\build.ps1\nbuild.ps1 downloads pre-built llama.cpp Vulkan DLLs and compiles dist\\janus.exe.\n2. Download a model\nPut a .gguf file in the models\\ folder. Use the included downloader:\ngo build -o dist\\modelget.exe .\\cmd\\modelget\n.\\dist\\modelget.exe -repo meta-llama/Llama-3.2-3B-Instruct -file Llama-3.2-3B-Instruct-Q8_0.gguf -out .\\models\\\nOr download any GGUF from Hugging Face.\n3. Configure\n\nEdit .env:\nINFERENCE_BACKEND=vulkan\nJANUS_MODEL_PATH=./models/Llama-3.2-3B-Instruct-Q8_0.gguf\nJANUS_MAX_TOKENS=4096\n\n\n\nVariable\nDefault\nMeaning\n\n\n\n\nINFERENCE_BACKEND\nvulkan\nvulkan, cpu, or openrouter\n\n\nJANUS_MODEL_PATH\n(required)\nPath to your .gguf file\n\n\nJANUS_GPU_LAYERS\n-1\n-1 = all layers on GPU, 0 = CPU only\n\n\nJANUS_VRAM_CEILING_MB\n9216\nVRAM budget hint (MiB)\n\n\nJANUS_MAX_TOKENS\n4096\nMax tokens per reply\n\n\nJANUS_LISTEN_ADDR\n127.0.0.1:8990\nBind address\n\n\n\n4. Run\n\nOpens http://127.0.0.1:8990 in your browser.\n5. Verify\ncurl http://127.0.0.1:8990/health\n\nQuick start (Linux / macOS)\ngit clone https://github.com/Vibra-Ingenn/Janus.git\ncd Janus\ngo mod tidy\ngo build -o dist/janus ./cmd/janus\ncp .env.example .env\n# edit .env \u2014 set JANUS_MODEL_PATH and INFERENCE_BACKEND=cpu if no Vulkan\n./dist/janus\nOn Linux you need libllama.so next to the binary or on LD_LIBRARY_PATH.\n\nOpenAI-compatible API\ncurl http://127.0.0.1:8990/v1/chat/completions \\\n -H \"Content-Type: application/json\" \\\n -d '{\n \"model\": \"local\",\n \"messages\": [{\"role\": \"user\", \"content\": \"Hello!\"}]\n }'\nBase URL: http://127.0.0.1:8990/v1\nEndpoints\n\n\n\nMethod\nPath\nDescription\n\n\n\n\nGET\n/health\nLiveness check (?deep=true for details)\n\n\nGET\n/v1/models\nModel list\n\n\nPOST\n/v1/chat/completions\nChat (streaming supported)\n\n\nPOST\n/models/load\nHot-swap model\n\n\nGET\n/models/list\nAvailable .gguf files\n\n\nGET\n/engine/status\nVRAM and backend info\n\n\n\nStreaming\ncurl http://127.0.0.1:8990/v1/chat/completions \\\n -H \"Content-Type: application/json\" \\\n -d '{\"model\":\"local\",\"stream\":true,\"messages\":[{\"role\":\"user\",\"content\":\"Tell me a joke\"}]}'\n\nConnect to Cursor / Cline / other clients\nBase URL: http://127.0.0.1:8990/v1\nAPI Key: (leave blank)\n\n\nPitfalls (learned the hard way)\n\n\n\nProblem\nWhat's going on\nFix\n\n\n\n\n\"It built but my changes aren't there\"\nOn Windows, Go can't overwrite a running .exe.\nStop all janus.exe in Task Manager, then rebuild.\n\n\n\"Address already in use\"\nA leftover process holds port 8990.\nTask Manager \u2192 end all janus.exe.\n\n\nServer starts, chat fails\nJANUS_MODEL_PATH wrong or no .gguf in models/.\nSet the path in .env, put the file in models/.\n\n\n\"local engine failed to start\"\nMissing llama.dll or GPU driver issue.\nRun .\\build.ps1. Update GPU drivers, or set INFERENCE_BACKEND=cpu.\n\n\nWrong URL\nDefault is http://127.0.0.1:8990, not 8080.\nBookmark 8990.\n\n\nFirst reply takes forever\nModel loading into VRAM \u2014 normal.\nWait 10\u201360s; smaller quants (Q4) load faster.\n\n\n\n\nProject layout\ncmd/janus/ Main server (OpenAI-compatible API)\ncmd/modelget/ Hugging Face model downloader\ninternal/engine/ llama.cpp Vulkan/CPU backend\ninternal/bridge/ DLL loader and FFI bindings\ninternal/singleton/ Single-instance guard\nmodels/ Put .gguf files here (not committed)\ndist/ janus.exe + llama.dll after build\n\n\nDevelopment\nGet-Process -Name \"janus\" -ErrorAction SilentlyContinue | Stop-Process -Force\ngo test ./...\ngo build -o dist\\janus.exe .\\cmd\\janus\n.\\dist\\janus.exe\nOr just .\\run.ps1. For a full build including llama DLLs, use .\\build.ps1.\nContributions welcome \u2014 see CONTRIBUTING.md.\n\nLicense\nMIT \u2014 see LICENSE." ], "output": "readability", "pwd": "/data/archive/1790892380.785368", "schema": "ArchiveResult", "start_ts": "2026-10-01T22:07:27.341193+00:00", "status": "succeeded" } ], "screenshot": [], "singlefile": [], "title": [ { "cmd": [ "curl", "--silent", "--location", "--compressed", "--max-time", "120", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://github.com/Vibra-Ingenn/Janus" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-10-01T22:07:27.333173+00:00", "index_texts": null, "output": "GitHub - Vibra-Ingenn/Janus: Janus is a API router for AI models written in Go and has a Vulkan Model runner \u00b7 GitHub", "pwd": "/data/archive/1790892380.785368", "schema": "ArchiveResult", "start_ts": "2026-10-01T22:07:27.242216+00:00", "status": "succeeded" } ], "wget": [ { "cmd": [ "wget", "--no-verbose", "--adjust-extension", "--convert-links", "--force-directories", "--span-hosts", "--no-parent", "-e", "robots=off", "--reject=js,mjs,woff,woff2,ttf,otf,eot,mp4,m4v,mov,webm,ogv,mkv,avi,flv", "--timeout=120", "--restrict-file-names=windows", "--page-requisites", "--user-agent=Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) wget/GNU Wget 1.21.3", "--compression=auto", "--timestamping", "https://github.com/Vibra-Ingenn/Janus" ], "cmd_version": "GNU Wget 1.21.3", "end_ts": "2026-10-01T22:07:27.232155+00:00", "index_texts": null, "output": "github.com/Vibra-Ingenn/Janus.html", "pwd": "/data/archive/1790892380.785368", "schema": "ArchiveResult", "start_ts": "2026-10-01T22:07:20.068691+00:00", "status": "succeeded" } ] }, "icons": null, "is_archived": true, "is_static": false, "latest": { "archive_org": null, "dom": null, "favicon": "favicon.ico", "git": null, "media": null, "pdf": null, "screenshot": null, "singlefile": null, "title": "GitHub - Vibra-Ingenn/Janus: Janus is a API router for AI models written in Go and has a Vulkan Model runner \u00b7 GitHub", "warc": null, "wget": "github.com/Vibra-Ingenn/Janus.html" }, "link_dir": "/data/archive/1790892380.785368", "newest_archive_date": "2026-10-01T22:07:30.390998+00:00", "num_failures": 0, "num_outputs": 6, "oldest_archive_date": "2026-10-01T22:07:19.056570+00:00", "path": "/Vibra-Ingenn/Janus", "schema": "Link", "scheme": "https", "snapshot_id": "214e5401-c334-4423-921f-ae3ebfba07ad", "sources": [ "/data/sources/1790892380-import.txt" ], "tags": null, "tags_str": "", "timestamp": "1790892380.785368", "title": "GitHub - Vibra-Ingenn/Janus: Janus is a API router for AI models written in Go and has a Vulkan Model runner \u00b7 GitHub", "updated": "2026-10-01T22:07:19.056150+00:00", "updated_date": "2026-10-01 22:07", "url": "https://github.com/Vibra-Ingenn/Janus" }