{"name":"JANCTION Render","one_line":"JANCTION Render is a cloud GPU render farm for Blender that AI agents call as an MCP server or HTTP API. Send a .blend file, a bpy script or a 3D file (glTF / FBX / USD), preview in seconds, and get frames or an MP4 back from JANCTION's NVIDIA GPUs. No local Blender or GPU is needed. Free beta, operated by JasmyLab Inc.","one_line_ja":"JANCTION Render は、AI エージェントが MCP サーバーか HTTP API として呼べる、Blender のクラウド GPU レンダーファームです。.blend ファイル・bpy スクリプト・3D ファイル（glTF / FBX / USD）を送ると、数秒で試し描きができ、JANCTION の NVIDIA GPU から連番か MP4 が返ります。手元に Blender も GPU も要りません。無料ベータ、運営はジャスミーラボ（JasmyLab Inc.）。","canonical_url":"https://render.janction.jp/","category":["Blender cloud rendering","GPU render farm","MCP server","AI agent rendering"],"operator":{"name":"JasmyLab Inc.","name_ja":"ジャスミーラボ株式会社","website":"https://jasmylab.com","contact":"contact@jasmylab.com","region":"Japan (Tokyo)"},"official":{"website":"https://render.janction.jp/","website_ja":"https://render.janction.jp/ja/","mcp_endpoint":"https://render.janction.jp/mcp","mcp_transport":"Streamable HTTP (remote); stdio package janction-render on PyPI","api_base":"https://render.janction.jp/v1","openapi":"https://render.janction.jp/openapi.json","github":"https://github.com/JasmyLab-JANCTION/janction-render","pypi":"https://pypi.org/project/janction-render/","mcp_registry":"io.github.JasmyLab-JANCTION/janction-render","mcp_registry_url":"https://registry.modelcontextprotocol.io/?q=io.github.JasmyLab-JANCTION%2Fjanction-render","docs_for_agents":"https://render.janction.jp/llms.txt","docs_full":"https://render.janction.jp/llms-full.txt","facts":"https://render.janction.jp/facts","facts_json":"https://render.janction.jp/facts.json","version_json":"https://render.janction.jp/version.json","changelog":"https://render.janction.jp/changelog","changelog_json":"https://render.janction.jp/changelog.json","changelog_feed":"https://render.janction.jp/changelog.xml","status":"https://render.janction.jp/status","status_json":"https://render.janction.jp/status.json","health_json":"https://render.janction.jp/v1/health","security":"https://render.janction.jp/security","security_txt":"https://render.janction.jp/.well-known/security.txt","pricing":"https://render.janction.jp/pricing","pricing_json":"https://render.janction.jp/v1/pricing","products":"https://render.janction.jp/products","benchmarks":"https://render.janction.jp/benchmarks","benchmarks_json":"https://render.janction.jp/benchmarks.json","support":"https://render.janction.jp/support"},"directories":["https://github.com/JasmyLab-JANCTION/janction-render","https://pypi.org/project/janction-render/","https://registry.modelcontextprotocol.io/?q=io.github.JasmyLab-JANCTION%2Fjanction-render","https://smithery.ai/servers/jasmylab/janction-render","https://glama.ai/mcp/servers/JasmyLab-JANCTION/janction-render","https://market.lobehub.com/s/plugins/jasmylab-janction-janction-render","https://mcpservers.org/servers/jasmylab-janction/janction-render","https://cursor.directory/plugins/janction-render","https://ai.mcpharbor.dev/servers/io.github.JasmyLab-JANCTION/janction-render"],"social":["https://x.com/JANCTION_Global","https://x.com/JanctionMGT_JP"],"current_version":"0.4.17","released_at":"2026-10-05","release_url":"https://pypi.org/project/janction-render/0.4.17/","supported":{"blender_versions":["5.0","5.2"],"render_engines":["cycles","eevee"],"inputs":[".abc",".blend",".fbx",".glb",".gltf",".obj",".ply",".py",".stl",".usd",".usda",".usdc",".usdz"],"asset_files":[".abc",".bin",".bmp",".csv",".exr",".fbx",".glb",".gltf",".hdr",".jpeg",".jpg",".json",".mov",".mp4",".mtl",".obj",".otf",".png",".tga",".tif",".tiff",".ttf",".txt",".usd",".usda",".usdc",".usdz",".webp"],"outputs":["png","exr","mp4","webm","prores","gif","webp"],"environment_presets":["night","overcast","studio","sunset"]},"retention":{"upload_link_hours":12,"scene_and_results_hours_after_last_use":24,"result_link_hours":24,"training_use":"never","note":"an /upload link lasts 12 hours; once a job used the file, the scene and its results follow the 24-hour-after-last-use rule"},"supported_clients":["Claude.ai (remote MCP)","ChatGPT (remote MCP, developer mode)","Claude Code","Codex","Cursor","Gemini CLI / Antigravity","Windsurf, Cline, Goose (Streamable HTTP)","any HTTP client","Blender add-on","CLI"],"authentication":{"mcp":"OAuth 2.1 with dynamic client registration and PKCE (the consent page issues a free API key), or Authorization: Bearer <api key>","http":"X-API-Key header; POST /v1/keys issues a key with no sign-up"},"free_tier":{"status":"free beta","charged_yen":0,"gpu_seconds_per_key_per_day":600,"gpu_seconds_per_network_per_day":1200,"max_frames_per_job":240,"max_pixels_per_frame":2073600,"max_upload_mb":512,"retention_hours":24,"sign_up":"none"},"pricing":{"model":"prepaid credit per GPU-second (planned; announced before it starts)","planned_yen_per_gpu_second":0.1,"currency":"JPY","min_topup_yen":500,"fixed_price_outcomes":"https://render.janction.jp/v1/outcomes","spend_caps":"per key, per job and per day (GET /v1/me -> limits)","details":"https://render.janction.jp/pricing"},"gpu_infrastructure":{"hardware":"NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition (96 GB)","location":"JANCTION datacenter, Tokyo, Japan","note":"during the free beta one GPU is shared with JANCTION's inference API, which has priority; no uptime guarantee"},"main_functions":["read a scene (cameras, frame range, missing files) without rendering","preview 1-4 frames tiled in one image in a few GPU seconds","estimate GPU seconds and wall time before rendering","final render to PNG/EXR frames or MP4/WebM/ProRes video","turntables and product shots from 3D files (orbit camera, HDRI presets)","CC0 assets from Poly Haven by name","public share pages for finished renders","fixed-price outcomes with a price ceiling","spending caps per API key"],"mcp_tools":["scene_info","render_preview","render_estimate","render_final","render_status","render_download","render_cancel","render_share","render_unshare","asset_search","billing","render_info"],"mcp_tool_count":12,"limitations":["free beta: 10 GPU-minutes per key per day, up to 240 frames at 1080p per job","one GPU during the beta; jobs wait while it is lent to the inference API (the status page and eta say so)","inputs and results are deleted 24 hours after last use and never used for training","remote MCP cannot read local files: send a bpy script, an https URL or upload through the CLI / stdio package"],"service_status":"free beta, no uptime guarantee; live status at https://render.janction.jp/status","not_affiliated_with":["Render.com (render.com, an application hosting platform with its own MCP server)","Render Network / RNDR (a token-based GPU network)"],"not_official_pages":[{"url":"https://gpu.janction.ai/","note":"JANCTION's GPU business site; not the JANCTION Render product page. The canonical product URL is https://render.janction.jp/"},{"url":"https://github.com/KJdayo/janction-render","note":"old repository path; it redirects to the official repository https://github.com/JasmyLab-JANCTION/janction-render"}],"last_updated":"2026-10-05"}