Video processing operations accessible through a single REST API.
Run any FFmpeg command via API. Sandboxed cloud execution, pay per job.
Add image and text watermarks to video and images programmatically.
Generate styled QR codes with 9 content types via API.
Extract captions with AI transcription or burn subtitles into video.
Automatically detect and blur faces and license plates in video.
Render video from a JSON timeline. Multi-track, transitions, text, shapes, and effects.
Generate an image from a text prompt, edit one from an instruction, or upscale one, on hosted diffusion models.
Shrink an image, video, or audio file to a target quality or size across 14 codecs.
Read codec, resolution, duration, and stream metadata from any media URL.
11 MCP tools for video processing from Claude, GPT, and other AI agents.
FFmpeg as a Service, the Compose API, animated captions, image generation, compression, media probing, and MCP integration are fully live. Run any FFmpeg command via the REST API, render video from a JSON timeline, generate an image from a prompt, or have AI agents drive video jobs through 11 MCP tools. Higher-level operations (managed watermarking, GDPR redaction, transcoding presets) are on the roadmap and accessible via FFmpeg in the meantime.
FFmpeg commands, compression and media probing run on FFmpeg, so you get predictable output and the same filters and codecs you'd use locally. Compose renders a JSON timeline on its own engine. Image generation, editing and upscaling run on GPU models. Higher-level features on the roadmap will be FFmpeg pipelines packaged with sensible defaults.
Yes. Email [email protected] with your use case. Coming-soon features ship faster when paying customers ask for them. We don't take requests for planned items unless they map to a real workflow you'd build today.
Yes. Anything FFmpeg can read, Rendobar can process. The live FFmpeg backend handles video, audio, image, and subtitle inputs. Watermarking, QR generation, and other coming-soon features will follow the same auto-detection model, but for now you can do all of it today via FFmpeg commands.
No. MCP tools wrap the same REST API. The 11 MCP tools (submit_job, get_job, list_jobs, probe_media, get_account, cancel_job, upload_media, complete_upload, get_upload, list_job_types, validate_compose) are 1:1 with REST endpoints. Use whichever fits your client. AI agents prefer MCP because tool-calling is native. Humans and apps use REST.
