JANCTION Render lets you render Blender on an NVIDIA RTX PRO 6000 Blackwell, a workstation GPU with 96 GB of memory, without buying one: JANCTION's own GPUs run Blender's Cycles and you pay 0.1 JPY per GPU-second from prepaid credit, after free time that needs no sign-up. Measured on the service: Blender's Classroom demo at 1920x1080, 128 samples, denoised, takes 8.93 GPU-seconds a frame (about 0.89 JPY) and the Barcelona Pavilion 10.55 (about 1.05 JPY). For a small scene a recent gaming GPU is about as fast per frame (a simple 1080p frame: 6.0 seconds on an RTX 5070 Ti laptop GPU, 6.7 here); the difference is the 96 GB of memory, which holds scenes that do not fit in 8 to 16 GB, and long animations rendered as parts in parallel, so the wait is shorter than one card rendering frame after frame. AI agents connect at https://render.janction.jp/mcp (remote MCP with OAuth; connecting creates a free key).
Updated 2026-10-14
What it does Your own .blend, bpy script or 3D file (glTF/GLB, FBX, USD, OBJ...) rendered to images or video on a cloud NVIDIA GPU, from an AI app, the terminal or the browser. The first preview needs no payment.
What it does not do It does not speed up the Blender viewport on your machine, does not take non-Blender video, simulations or general GPU jobs, and an AI that is not connected cannot render on its own.
Try your own file for freeBuild it with an AI and try
Browser only; no connector or CLI needed. A 100 JPY welcome credit for a new key (14 days, no card; one per network every 30 days, further keys on the same network share what is left), and 400 JPY more for saving a card that is not charged (once per card); after that, previews stay free up to 2 GPU-minutes a day and finals cost 0.1 JPY per GPU-second.
| scene | settings | GPU-seconds per frame | price per frame |
|---|---|---|---|
| Blender's Classroom demo | 1920x1080, Cycles, 128 samples, denoised | 8.93 | about 0.89 JPY |
| Barcelona Pavilion demo | 1920x1080, Cycles, 128 samples, denoised | 10.55 | about 1.05 JPY |
| a rotating cube (48 frames) | 1920x1080, Cycles, 128 samples | 6.7 | about 0.67 JPY |
| a 4-frame preview | 640x360 tiles, 16 samples | about 4 for the sheet | 0 (previews are free up to 2 GPU-minutes a day) |
GPU-seconds include starting Blender and reading the scene. How each number was measured, with the files: Blender render cost and benchmarks.
| JANCTION Render | buy a workstation card | rent a cloud GPU by the hour | a render farm | |
|---|---|---|---|---|
| Start | no sign-up; from the browser, an AI app over MCP, the CLI or the API | buy and install the card | an account, a machine image, Blender installed by you | an account and usually a desktop client |
| You pay for | GPU-seconds actually used, from 1 JPY | the card, up front | every hour the machine is on, including setup and idle time | their unit (node-hour, GHz-hour, GPU-hour) |
| Preview loop | 1 to 4 frames in seconds, with a verdict and fix code an AI can act on | your own Blender | what you build | test renders through their client |
| Fits | still images, product shots, turntables and animations up to 3,000 frames per instruction; scenes and AI-made models that need a lot of GPU memory | daily heavy use | custom pipelines you maintain | long sequences with studio support |
Service-by-service comparisons with each company's published prices: render farms for AI agents, render farm APIs, RebusFarm and the studio farms, SheepIt, all comparisons.
| from | how |
|---|---|
| A file on this computer (.blend, 3D file) | Drop it at /upload to get a 12-hour link (no sign-up), then tell the AI "Render this file: <link>". From inside Blender use the add-on; from a terminal the CLI uploads directly |
| Claude.ai (web / desktop / mobile) | Settings → Connectors → Add custom connector → https://render.janction.jp/mcp. "Connect" opens a page that creates a free key |
| ChatGPT | Settings → Connectors → Advanced → Developer mode → Create → https://render.janction.jp/mcp (OAuth) |
| Claude Code | claude mcp add --transport http janction-render https://render.janction.jp/mcp, then /mcp to authenticate |
| Codex | codex mcp add janction-render --url https://render.janction.jp/mcp |
| Cursor / VS Code | Add to Cursor Add to VS Code The button opens the editor; confirm and it is added. Manual: Streamable HTTP at https://render.janction.jp/mcp |
| Antigravity CLI / Gemini CLI | agy plugin install https://github.com/JasmyLab-JANCTION/janction-render or gemini extensions install https://github.com/JasmyLab-JANCTION/janction-render (the MCP server and the workflow skill) |
| Grok / Perplexity / Le Chat | Add a custom (remote MCP) connector with https://render.janction.jp/mcp in each app's Connectors settings (Grok: New Connector → Custom; Perplexity: Pro / Max / Enterprise; Le Chat: workspace admin) |
| Notion (Notion Agent) | Business / Enterprise: Settings → Connections → Discover → Add Custom MCP → https://render.janction.jp/mcp, then sign in (the button shows when a workspace admin allows custom MCP servers) |
| Windsurf, Cline, Goose, other MCP clients | Streamable HTTP at https://render.janction.jp/mcp; OAuth or Authorization: Bearer <api key> |
| Claude Desktop and other stdio-only clients (npm) | npx -y janction-render (in the client's mcpServers: {"command": "npx", "args": ["-y", "janction-render"]}). The npm package janction-render bridges stdio to https://render.janction.jp/mcp with OAuth; --key jr_… for a key you have, --local for the Python stdio server through uv |
| Terminal (stdio, sends local files) | claude mcp add janction-render -e JANCTION_RENDER_SERVER=https://render.janction.jp -- uvx --from janction-render janction-render-mcp |
| Inside Blender (for people) | The add-on; extension repository https://render.janction.jp/extensions/index.json. Maya / Houdini / Cinema 4D: submit tools |
For small scenes, not by much: a simple 1080p frame took 6.0 seconds on an RTX 5070 Ti laptop GPU and 6.7 here, because most of a short frame is fixed work. It matters for scenes that need more memory than a gaming card has, and for animations, which are rendered as parts in parallel.
0.1 JPY per GPU-second, which is 360 JPY per GPU-hour, billed by the second from prepaid credit after the free part. A job pays only for the GPU-seconds it actually uses.
No. Jobs run on JANCTION's own GPUs (NVIDIA RTX PRO 6000 Blackwell). When they are busy a job waits in a queue, and the estimate says how long.
JANCTION Render v0.4.115 · JasmyLab Inc. · Updated 2026-10-14 · Tested with Blender 5.0 / 5.2, JANCTION Render 0.4.115 · 日本語