llms.txt
+80 −75
https://varosity.ai/llms.txt
− Varosity routes agent-initiated generation requests across providers including fal.ai, Runway, Luma, ElevenLabs, Suno, OpenAI, Replicate, Pika, Hailuo, and others. The customer can use their own provider accounts (zero markup, billed directly by the provider) or Varosity credits (no provider accounts needed; about 5% over provider list price). The MCP server, REST API, and CLI all use the same `vsk_` API key.− Per-agent setup (Claude Code, Cursor, Grok Bot, Gemini CLI, Manus, Perplexity): https://varosity.ai/connect− Grok Bot guide and standing-job examples: https://varosity.ai/grok-bot− Install (remote, Streamable HTTP):− POST https://varosity.ai/api/v1/credits/checkout — Stripe checkout link that adds credits ({ packId: "starter" | "pro" }). Pay it yourself or give it to the user. MCP: buy_credits.− No key yet? POST https://varosity.ai/api/v1/agent/connect { agentName } (no key needed), show your user verification_uri_complete, then POST { request_id, poll_token } to https://varosity.ai/api/v1/agent/connect/poll every 5 s until it returns api_key. Or send your user to https://varosity.ai/agent/key and they paste the key back.− - [5-minute first render](https://varosity.ai/docs/quick-start) — Describe a video, pick a frame, get a clip — no provider keys needed.− - [Agent mode (Claude / Cursor / MCP)](https://varosity.ai/docs/agent-mode) — Drive Varosity from any MCP host: 40+ tools, bearer auth, JSON-RPC over Streamable HTTP.− - [Avatar on any background](https://varosity.ai/docs/avatar-on-any-background) — Composite a talking head on a Veo / Kling / Runway background.− - [Background music + auto-ducking](https://varosity.ai/docs/background-music-and-ducking) — Add a soundtrack to a project; voice ducks the music automatically.− - [BYOK setup per provider](https://varosity.ai/docs/byok-setup) — Where to get each provider's key + how to add it to Varosity.− - [Chaining clips across models](https://varosity.ai/docs/multi-shot-storyboard) — Use the right model for each shot type, then stitch them into one MP4.− - [Managing renders + costs](https://varosity.ai/docs/managing-renders) — Where renders live, how to download, and how to keep BYOK costs under control.− - [Prompting cheatsheet](https://varosity.ai/docs/prompting-cheatsheet) — Camera, lighting, and motion language that lands consistently across models.− - [Using the CLI](https://varosity.ai/docs/using-the-cli) — Install @varosity/cli; render videos and music from the terminal.− - [Voice over your video](https://varosity.ai/docs/voice-over-your-video) — Add narration with ElevenLabs, OpenAI TTS, or Deepgram.− BYOK is zero markup — customers pay providers directly. Varosity credits cover calls with no provider accounts needed (about 5% over provider list price), sold in packs starting at $10. Signing up is free and most new accounts get a $1.50 starter credit. Pricing page: https://varosity.ai/pricing− Self-updating skill files for Hermes, Claude Code, Claude Desktop, and other MCP hosts.− - [custom-video-generation](https://varosity.ai/api/v1/skills/custom-video-generation) — On-demand single-video pipeline: reference image pre-flight → render → raw delivery.− - `veo-3.1` (fal) — best lip sync, native 4K, synced audio, talking heads. ~$0.15/s, max 8s.− - `veo-3.1-quality` (fal) — state-of-the-art fidelity, best lip sync, native 48kHz audio, native 4K. ~$0.4/s, max 8s.− - `kling-3.0` (fal) — multi-shot consistency, cinematic motion, dialogue close-ups, 4K. ~$0.17/s, max 10s.− - `grok-imagine-video` (fal) — xAI's Grok Imagine — a low-priced frontier video model with native synchronized audio ($0.07/s, sound included). Flexible 1–15s clips, text-to-video and image-to-video.. ~$0.07/s, max 15s.− - `seedance-4.5` (fal) — unified audio-video generation, multi-shot from one prompt, phoneme-level lip-sync. ~$0.14/s, max 12s.− - `seedance-2.0` (fal) — ByteDance's flagship video model — best-in-class motion, physics and cinematic camera control with native synced audio, up to 15s. Supports start+end keyframe transitions.. ~$0.3024/s, max 15s.− - `seedance-2.0-fast` (fal) — The budget tier of Seedance 2.0 — near-flagship motion and physics at lower cost and latency, with native synced audio and start+end keyframes. Up to 15s.. ~$0.2419/s, max 15s.− - `omnihuman` (fal) — Drives a talking avatar from one photo + an audio clip. Drop on any shot in a brand agent's storyboard.. ~$0.17/s, max 60s.− - `fal-live-portrait` (fal) — Recreate a performer's exact head movement, facial expressions and lip motion on a DIFFERENT person. Supply a driving video (referenceVideoUrl) + a target face (referenceImageUrl). Face/head transfer; pair with a cloned voice for the full effect.. ~$0.05/s, max 60s.− - `fal-wan-animate` (fal) — Recreate a performer's FULL-BODY movements (and face) on a DIFFERENT character. Supply a driving video (referenceVideoUrl) + a character image (referenceImageUrl). Wan 2.2 Animate via fal — the full-body version of performance transfer, live today.. ~$0.16/s, max 120s.− - `kling-3.0-replicate` (replicate) — multi-shot consistency, cinematic motion, alternate route when fal is busy. ~$0.34/s, max 10s.− - `happyhorse-1.0` (fal) — Alibaba's HappyHorse 1.0 — the #1-ranked AI video model on Artificial Analysis (text- and image-to-video). Joint audio+video, multilingual lip-sync, up to 1080p.. ~$0.14/s, max 15s.− - `luma-ray-3-fal` (fal) — high-fidelity motion, strong realism. ~$0.12/s, max 10s.− - `runway-gen-4.5-replicate` (replicate) — camera control, motion brush, scene consistency, prompt adherence. ~$0.12/s, max 10s.− - `hailuo-02` (hailuo) — physics realism, complex motion, long prompts. ~$0.11/s, max 10s.− - `muapi-wan-effects` (muapi) — Apply Cakeify, VHS, Samurai, Film Noir and 20+ other AI effects to images. Billed in Varosity Credits.. ~$0.06/s, max 10s.− - `muapi-latentsync` (muapi) — Sync lip movements to any audio track on an existing video. Billed in Varosity Credits.. ~$0.05/s, max 60s.− - `muapi-wan-t2v` (muapi) — WAN 2.1 text-to-video via muapi — platform-funded fallback when WaveSpeed is unavailable. No provider key needed.. ~$0.065/s, max 10s.− - `heygen-avatar-4` (heygen) — HeyGen's flagship talking-avatar (Digital Twin) on the v3 API — the highest-quality lip-sync available. Drives a pre-trained avatar look with a script (or your own audio), with the Avatar IV motion engine, voice emotion/speed/pitch, captions, and custom backgrounds.. ~$0.08/s, max 300s.− - `heygen-photo-avatar` (heygen) — Turn a single photo into a talking presenter with HeyGen's Avatar IV engine. Pass a photo URL + script and HeyGen animates the face with natural motion — no pre-training required. Control motion via a prompt and expressiveness level.. ~$0.08/s, max 300s.− - `heygen-cinematic` (heygen) — Prompt-driven cinematic shots featuring your avatar — describe a scene (camera, setting, action) and HeyGen renders a documentary-style clip from 1–3 avatar looks, optionally guided by reference clips/images.. ~$0.1/s, max 15s.+ Varosity routes agent-initiated generation requests across providers including fal.ai, Runway, Luma, ElevenLabs, Suno, OpenAI, Replicate, Pika, Hailuo, and others. The customer can use their own provider accounts (billed directly by the provider) or Varosity credits (no provider accounts needed). Either way every model costs its provider's list price, with no markup. The MCP server, REST API, and CLI all use the same `vsk_` API key.+ Per-agent setup (Claude Code, Cursor, Grok Bot, Gemini CLI, Codex, Manus, Perplexity): https://varosity.ai/connect+ Guides with standing-job examples: https://varosity.ai/claude-code, https://varosity.ai/cursor, https://varosity.ai/gemini-cli, https://varosity.ai/codex, https://varosity.ai/grok-bot+ No key needed: add the URL and sign in when the host asks. The token has a $5 daily cap and fewer tools than a vsk_ key (no brand agents, storyboards, transcription, webhook changes or asset governance).+ Claude Code: claude mcp add --transport http varosity https://varosity.ai/api/mcp — then /mcp → varosity → Authenticate+ Install with a key (remote, Streamable HTTP):+ What a call was charged, in cents: image results carry cost_cents and billing_mode; finished jobs carry costCents.+ Tell your user each time you use Varosity: one short line with the model and what it cost, from the result (e.g. "Made with Kling 3.0 on Varosity — 84¢"). On their own provider key there is no Varosity charge; say so.+ POST https://varosity.ai/api/v1/credits/topup?amountUsd=10 — Pay the request itself (Machine Payments Protocol, HTTP 402): card via a Link agent wallet or USDC; no checkout page. Send your key as X-API-Key (Authorization carries the payment), e.g. `npx @stripe/link-cli mpp pay <url> -X POST -H "X-API-Key: vsk_..."`.+ POST https://varosity.ai/api/v1/pay/generate — Pay for ONE image or video, no credits. Body {"type":"image"|"video","model":"kling-3.0","prompt":"…"}. The 402 names job_id, model and price (credits price, min $0.50, max $50); pay and send the same request again — that runs the job. Images return output_url; videos poll GET /api/v1/jobs/{job_id}. Refunded if it fails.+ POST https://varosity.ai/api/v1/credits/checkout — Stripe checkout link that adds credits ({ amountUsd: 5–500 } or { packId: "starter" | "pro" }). Pay it yourself or give it to the user. MCP: buy_credits.+ No key yet? POST https://varosity.ai/api/v1/agent/connect { agentName } (no key needed; your user approves once and picks a daily limit of $5, $25 or $100; the key expires in 90 days; apps that sign in with OAuth get $5 a day), show your user verification_uri_complete (ask them to reply once approved), then POST { request_id, poll_token } to https://varosity.ai/api/v1/agent/connect/poll every "interval" seconds (from the connect response; on slow_down, wait longer). Pending polls return HTTP 4+ - [5-minute first render](https://varosity.ai/docs/quick-start) — Your first AI video in five minutes: describe it, pick one of three frames, get a clip. Runs on Varosity credits, so no provider keys are needed.+ - [Agent mode (Claude / Cursor / MCP)](https://varosity.ai/docs/agent-mode) — Drive Varosity from any MCP host: 46 tools (29 for apps that sign in with OAuth), bearer auth, JSON-RPC over Streamable HTTP.+ - [Avatar on any background](https://varosity.ai/docs/avatar-on-any-background) — Put an OmniHuman talking-head avatar over a background clip from Kling, Runway or another model, picture-in-picture, in one render.+ - [Background music](https://varosity.ai/docs/background-music-and-ducking) — Generate a soundtrack with ElevenLabs Music, Lyria or Suno and lay it under your video with the stitch step. Auto-ducking under speech is not available yet.+ - [BYOK setup per provider](https://varosity.ai/docs/byok-setup) — Where to get an API key for fal.ai, Replicate, ElevenLabs, OpenAI, Google, Fish Audio and Cartesia, and how to add, test and rotate it in Varosity.+ - [Chaining clips across models](https://varosity.ai/docs/multi-shot-storyboard) — Pick the right model for each shot — dialogue, action, b-roll — then stitch the clips into one MP4 with Varosity's stitch step.+ - [Managing renders + costs](https://varosity.ai/docs/managing-renders) — Where renders live, how long provider video links last, how to download them, and how to keep spend under control with caps and your own keys.+ - [Prompting cheatsheet](https://varosity.ai/docs/prompting-cheatsheet) — Camera moves, lighting and motion phrases that work the same way across Kling, Veo, Seedance, Runway and other video models, with examples.+ - [Using the CLI](https://varosity.ai/docs/using-the-cli) — Install @varosity/cli, pair it with your account, and render videos and music from the terminal or a script.+ - [Voice over your video](https://varosity.ai/docs/voice-over-your-video) — Add narration to a video with ElevenLabs, OpenAI TTS, Deepgram, Cartesia or Fish Audio: pick a voice, generate the line, and mix it into the final cut.+ Every model is sold at its provider's list price, with no markup — on Varosity credits or on the customer's own provider keys. Credits come from monthly plans (Basic $12 → 1,200 credits, Pro $29 → 3,500, Creator $125 → 17,000; 1 credit = 1¢; plan credits expire monthly; yearly about 20% off) or one-off top-ups from $5 that never expire. Most verified accounts without a plan get 10 free credits a day (a limited number are given out each day; they expire at midnight UTC). Signing up is free and mo+ Skill playbooks for Hermes, Claude Code, Claude Desktop, and other MCP hosts — read the latest with refresh_skills { slug } or GET /api/v1/skills/{slug}.+ - [custom-video-generation](https://varosity.ai/api/v1/skills/custom-video-generation) — On-demand single-video pipeline: optional reference-image pick → render → raw delivery.+ - `veo-3.1` (fal) — best lip sync, 4K on your own fal key (720p/1080p on Credits), synced audio, talking heads. ~$0.15/s, max 8s.+ - `veo-3.1-quality` (fal) — state-of-the-art fidelity, best lip sync, native 48kHz audio, 4K on your own fal key. ~$0.4/s, max 8s.+ - `kling-3.0` (fal) — multi-shot consistency, cinematic motion, dialogue close-ups. ~$0.168/s, max 10s.+ - `grok-imagine-video` (fal) — xAI's Grok Imagine — a low-priced frontier video model with native synchronized audio ($0.07/s, sound included). Flexible 1–15s clips, text-to-video and image-to-video. ~$0.07/s, max 15s.+ - `seedance-1.0-pro` (fal) — multi-shot from one prompt, 1080p by default, single takes up to 12s. ~$0.1215/s, max 12s.+ - `seedance-2.0` (fal) — ByteDance's flagship video model — best-in-class motion, physics and cinematic camera control with native synced audio, up to 15s. Supports start+end keyframe transitions. ~$0.3024/s, max 15s.+ - `seedance-2.0-fast` (fal) — The budget tier of Seedance 2.0 — near-flagship motion and physics at lower cost and latency, with native synced audio and start+end keyframes. Up to 15s. ~$0.2419/s, max 15s.+ - `omnihuman` (fal) — Drives a talking avatar from one photo + an audio clip. Drop on any shot in a brand agent's storyboard. ~$0.16/s, max 60s.+ - `fal-live-portrait` (fal) — Recreate a performer's exact head movement, facial expressions and lip motion on a DIFFERENT person. Supply a driving video (referenceVideoUrl) + a target face (referenceImageUrl). Face/head transfer; pair with a cloned voice for the full effect. ~$0.05/s, max 60s.+ - `fal-wan-animate` (fal) — Recreate a performer's FULL-BODY movements (and face) on a DIFFERENT character. Supply a driving video (referenceVideoUrl) + a character image (referenceImageUrl). Wan 2.2 Animate via fal — the full-body version of performance transfer, live today. ~$0.16/s, max 120s.+ - `kling-3.0-replicate` (replicate) — multi-shot consistency, cinematic motion, alternate route when fal is busy. ~$0.336/s, max 10s.+ - `happyhorse-1.0` (fal) — Alibaba's HappyHorse 1.0 — the #1-ranked AI video model on Artificial Analysis (text- and image-to-video). Joint audio+video, multilingual lip-sync, up to 1080p. ~$0.14/s, max 15s.+ - `luma-ray-3-fal` (fal) — high-fidelity motion, strong realism. ~$0.1/s, max 10s.+ - `runway-gen-4.5-replicate` (replicate) — camera control, scene consistency, prompt adherence. ~$0.12/s, max 10s.+ - `muapi-wan-effects` (muapi) — Apply Cakeify, VHS, Samurai, Film Noir and 20+ other AI effects to images. Billed in Varosity Credits. ~$0.06/s, max 10s.
Showing the first 40 of each.
