[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f2qqiqvn5f1k1y":3},{"code":4,"message":5,"data":6,"timestamp":26},10000,"success",{"full_name":7,"position":8,"description":9,"language":10,"stars":11,"forks":12,"stars_gained":13,"age_days":14,"hn_threads":15,"one_liner":16,"summary_md":17,"topic":18,"github_topics":21,"license":22,"homepage":23,"updated_at":24,"history":25},"FujiwaraChoki\u002Fsupoclip",16,"An open-source OpusClip alternative","Python",1841,440,341,453,[],"Turns long videos and stream VODs into short vertical clips, for creators who don't want watermark or seat-based limits.","## The problem\n\nPodcasters and streamers need a steady flow of short vertical clips for TikTok, Reels and Shorts, and the work is slow: scrubbing a long VOD, choosing moments, cutting to 9:16, burning in captions. Hosted tools charge per minute of footage, often watermark exports, and keep your source files on their servers. SupoClip runs the same pipeline on your own hardware, or on its hosted and iOS versions if you don't want to run infrastructure.\n\n## What's different\n\nCompared with OpusClip and similar SaaS clippers, the main difference is where the pipeline runs: Docker Compose on your machine, unlimited processing bounded only by your hardware, no burned-in branding. It also exposes the pipeline as a REST API and an MCP server, so agents and scripts can submit jobs — most hosted clippers only give you a web UI. Internally, it chains an LLM for segment selection and virality scoring, face detection for 9:16 cropping, and word-level timestamps for subtitles, and it can swap providers (Gemini, GPT, Claude, or local Ollama) as well as run Whisper locally. That provider flexibility is rarer in this category and matters if you can't send footage to third parties.\n\n## Who it's for\n\nCreators and small teams with their own server or a beefy workstation who publish several clips a week and care about ownership and per-minute cost. It's also a reasonable base for developers who want to fork a clipping pipeline rather than build one. Skip it if you only clip occasionally — the hosted version is the sane option there — or if you need a mature product with support SLAs; this is a young project. Teams with strict data-residency needs should note that transcription defaults to AssemblyAI, so a fully offline setup means configuring Whisper and Ollama yourself.\n\n## Cost of adoption\n\nYou need Docker, an AssemblyAI key for transcription, and at least one LLM provider key; local Whisper and Ollama are possible but take extra networking setup. The stack is FastAPI plus an ARQ worker on the backend and Next.js on the frontend, so expect a multi-service deployment, not a single binary. The license is AGPL-3.0, which is fine for self-hosting but may matter if you plan to offer it as a hosted service. The README is unusually detailed — setup, configuration, architecture, troubleshooting are all documented — but the project is new and the frontend port changed recently, so expect churn. The start command is `docker-compose up -d`.\n\n## Verdict\n\nSupoClip is a genuine open-source answer to OpusClip rather than a thin wrapper: the pipeline, editor, publishing and the MCP\u002FREST surface are all there. It is also brand-new — about 15 months old, and its first tracked appearance is a +339-star week on the Python board in October 2026, with no Hacker News discussion. That pattern looks like a momentum spike rather than a settled project, so evaluate it as fast-moving code you may need to patch yourself. Adopt it if you can amortize the Docker setup and want to own the pipeline; don't if you need something to just work without touching configuration.",{"slug":19,"name":20},"media","Media & Design",[],"AGPL-3.0","https:\u002F\u002Fwww.supoclip.com\u002F","2026-10-24T22:36:45Z",[],1792898567]