# varg

## Docs

- [varg](https://docs.varg.ai/index.md): AI video generation platform. Install the skill, prompt your agent, get videos.
- [Quickstart](https://docs.varg.ai/quickstart.md): Get up and running with varg in 2 minutes
- [Authentication](https://docs.varg.ai/authentication.md): API keys, login, credentials, and billing
- [Bring Your Own Keys](https://docs.varg.ai/byok.md): Using your own provider API keys with varg
- [Models](https://docs.varg.ai/models/index.md): All AI models available through varg — video, image, lipsync, upscale, speech, music, and transcription
- [Seedance 2](https://docs.varg.ai/models/video/seedance-2.md): ByteDance's flagship video generation model — text-to-video up to 4k, image-to-video, and reference-to-video with multi-modal @-references
- [Kling V3](https://docs.varg.ai/models/video/kling-v3.md): Best overall video generation model — flexible duration, high quality text-to-video and image-to-video
- [Sora 2](https://docs.varg.ai/models/video/sora-2.md): OpenAI's video generation model — cinematic quality with text-to-video, image-to-video, and video remix
- [Wan 2.5](https://docs.varg.ai/models/video/wan-2.md): Budget-friendly video generation with strong character motion and audio support
- [Minimax](https://docs.varg.ai/models/video/minimax.md): Versatile video generation with good motion quality at moderate cost
- [LTX](https://docs.varg.ai/models/video/ltx.md): Fastest and cheapest video generation model — supports native audio and keyframe control
- [Grok Imagine Video](https://docs.varg.ai/models/video/grok-imagine-video.md): xAI's video generation model with native audio, video editing, and multiple resolutions
- [Kling Legacy](https://docs.varg.ai/models/video/kling-legacy.md): Previous generation Kling video models — V2.6, V2.5, V2.1, V2
- [Nano Banana (Gemini)](https://docs.varg.ai/models/image/nano-banana.md): Google Gemini-powered image generation and editing — best default image model with reference-based editing
- [Flux](https://docs.varg.ai/models/image/flux.md): Black Forest Labs image generation — three tiers from fast to premium quality
- [Phota](https://docs.varg.ai/models/image/phota.md): Identity-preserving image generation with profile-based character control
- [Grok Imagine Image](https://docs.varg.ai/models/image/grok-imagine-image.md): xAI's cheapest image generation and editing model
- [Recraft](https://docs.varg.ai/models/image/recraft.md): Purpose-built for graphic design — icons, illustrations, brand assets with precise color control
- [Qwen Image](https://docs.varg.ai/models/image/qwen.md): Alibaba's image models — multi-angle generation and unified image generation/editing
- [Seedream](https://docs.varg.ai/models/image/seedream.md): ByteDance's image models — Seedream 5 Lite for text-to-image with reference images, Seedream V4.5 for precise image editing
- [Soul](https://docs.varg.ai/models/image/soul.md): Character-focused image generation with 80+ style presets — ideal for consistent characters across scenes
- [VEED Fabric](https://docs.varg.ai/models/lipsync/veed-fabric.md): Fast talking head generation from a single image and audio — simplest lipsync pipeline
- [Sync Lipsync](https://docs.varg.ai/models/lipsync/sync.md): High-quality lip synchronization — apply speech audio to existing video
- [OmniHuman](https://docs.varg.ai/models/lipsync/omnihuman.md): ByteDance's full-body human animation from a single image and audio
- [Image Upscale](https://docs.varg.ai/models/upscale/image-upscale.md): Enhance and upscale images with 6 models at different quality-cost tradeoffs
- [Video Upscale](https://docs.varg.ai/models/upscale/video-upscale.md): Enhance video resolution with 4 models from budget to premium quality
- [ElevenLabs Speech](https://docs.varg.ai/models/speech/elevenlabs.md): Industry-leading text-to-speech with 21+ voices, multiple languages, and 7 model variants
- [Music Generation](https://docs.varg.ai/models/music/music.md): AI music generation for video soundtracks, background music, and jingles
- [Whisper Transcription](https://docs.varg.ai/models/transcription/whisper.md): Audio-to-text transcription with OpenAI Whisper — fal and Groq variants
- [SDK Overview](https://docs.varg.ai/sdk/index.md): vargai SDK for programmatic AI video generation
- [Components](https://docs.varg.ai/sdk/components.md): All varg JSX components and their props
- [AI Models](https://docs.varg.ai/sdk/models.md): Quick reference for all supported models — see per-model pages for detailed docs
- [CLI Reference](https://docs.varg.ai/sdk/cli.md): varg command-line interface
- [Templates](https://docs.varg.ai/templates/index.md): Copy-paste templates for common video types
- [Slideshow Template](https://docs.varg.ai/templates/slideshow.md): Create image slideshows with transitions, zoom effects, and music
- [Talking Character Template](https://docs.varg.ai/templates/talking-character.md): Create AI talking head videos with lipsync, voiceover, and captions
- [Transformation Template](https://docs.varg.ai/templates/transformation.md): Create before/after comparison videos for fitness, makeover, and product content
- [MCP Server](https://docs.varg.ai/mcp-server.md): Connect Claude, ChatGPT, VS Code, Cursor, and other AI tools to varg.ai
- [AI Agent Context](https://docs.varg.ai/ai-agents/index.md): Complete context for AI assistants to help users create videos with varg
- [Claude Code Setup](https://docs.varg.ai/ai-agents/claude-code.md): Configure Claude Code for video generation with varg
- [Cursor Setup](https://docs.varg.ai/ai-agents/cursor.md): Configure Cursor for video generation with varg
- [varg vs Remotion vs Hyperframes](https://docs.varg.ai/comparisons/varg-vs-remotion-vs-hyperframes.md): How varg, Remotion, and Hyperframes compare for programmatic video — and why they solve different problems.
- [varg API](https://docs.varg.ai/api/index.md): Unified REST API for AI media generation at api.varg.ai/v2
- [Folders](https://docs.varg.ai/api/folders.md): Group files into folders and file generated output automatically
- [Render API](https://docs.varg.ai/render-api/index.md): Submit TSX code and get rendered videos — zero dependencies, just curl
- [Generate an image](https://docs.varg.ai/api-reference/generation/generate-an-image.md): Create an image generation job. Also handles image editing (send a source image in `files` with an edit-capable model like `nano_banana_pro/edit`) and image upscaling (`clarity_upscaler`, `aura_sr`, `topaz`, ...).
- [Generate a video](https://docs.varg.ai/api-reference/generation/generate-a-video.md): Create a video generation job. Handles text-to-video, image-to-video (send a start frame in `files`), lipsync (`sync_v2`, `veed_fabric_1.0`, `omnihuman_v1.5`, ...), video editing (`grok_imagine_edit`), avatars (`heygen_avatar`), and video upscaling (`topaz_video`, `seedvr_video`).
- [Generate speech (TTS)](https://docs.varg.ai/api-reference/generation/generate-speech-tts.md): Convert text to speech. ElevenLabs models (`eleven_v3`, `eleven_turbo_v2_5`, `eleven_multilingual_v2`, ...).
- [Generate music](https://docs.varg.ai/api-reference/generation/generate-music.md): Generate a music track from a text prompt (`music_v1`, `eleven_music`).
- [Transcribe audio](https://docs.varg.ai/api-reference/generation/transcribe-audio.md): Transcribe an audio file by URL (`whisper`, `groq_whisper_large_v3_turbo`, ...).
- [Run an ffmpeg operation](https://docs.varg.ai/api-reference/generation/run-an-ffmpeg-operation.md): Media processing via ffmpeg. One endpoint, four operation modes selected by the `model` field:
- [Render TSX to video](https://docs.varg.ai/api-reference/generation/render-tsx-to-video.md): Submit varg TSX code and get back a rendered video (or image frames). No local setup needed — the render runs in the cloud, generating AI assets and stitching them together.
- [Run a custom pipeline](https://docs.varg.ai/api-reference/generation/run-a-custom-pipeline.md): Run a registered multi-step pipeline (e.g. `v2vStitch`). Pipelines are access-gated per account — contact varg to get access to a pipeline. The `model` field is the pipeline name; remaining fields are defined by the pipeline's own input schema (discover it via `GET /tools/pipeline?model=<name>`).
- [List tools](https://docs.varg.ai/api-reference/tools/list-tools.md): List every registered tool (capability) with a short description.
- [Get tool metadata](https://docs.varg.ai/api-reference/tools/get-tool-metadata.md): Full metadata for one tool: JSON Schema for its input, output shape, and a worked example. Pass `?model=<model>` to enrich `provider_options` with the explicit per-provider schema for that model (instead of the generic passthrough bag).
- [Call a tool generically](https://docs.varg.ai/api-reference/tools/call-a-tool-generically.md): Generic equivalent of the sugar routes — `POST /tools/image/call` is the same as `POST /image`. Useful for dynamic clients that discover tools at runtime via `GET /tools`.
- [List jobs](https://docs.varg.ai/api-reference/jobs/list-jobs.md): List the caller's jobs, newest first.
- [Get a job](https://docs.varg.ai/api-reference/jobs/get-a-job.md): Full job view. Poll this until `status` is terminal (`completed`, `failed`, or `cancelled`), then read `output.outputs[0].url`.
- [Get job status (lightweight)](https://docs.varg.ai/api-reference/jobs/get-job-status-lightweight.md): Minimal status projection — cheaper to poll than the full job view.
- [Get job pricing breakdown](https://docs.varg.ai/api-reference/jobs/get-job-pricing-breakdown.md)
- [Refresh a job](https://docs.varg.ai/api-reference/jobs/refresh-a-job.md): Force a re-poll of the provider and mirror the latest state onto the job.
- [Cancel a job](https://docs.varg.ai/api-reference/jobs/cancel-a-job.md): Abort a running job (best-effort at the provider). Returns 409 if the job is already terminal.
- [Retry a failed job](https://docs.varg.ai/api-reference/jobs/retry-a-failed-job.md)
- [Upload a file](https://docs.varg.ai/api-reference/files/upload-a-file.md): Upload a file as a raw binary body (max 200 MB). Set `Content-Type` to the file's media type; when it's missing or generic, varg sniffs the type from magic bytes.
- [List files](https://docs.varg.ai/api-reference/files/list-files.md): List the caller's files. Filter by creation criteria (`?tool=speech`, `?model=eleven_turbo_v2`, `?q=hello`), by kind (`?kind=image|video|audio` for media type, `?kind=upload|generated` for origin), or by date range (`?from=`, `?to=`, ISO 8601).
- [Import a file by URL](https://docs.varg.ai/api-reference/files/import-a-file-by-url.md): Register a file from a URL. varg-hosted URLs (`*.varg.ai`) are registered without re-uploading; external URLs are streamed to varg storage. SSRF-protected (private IPs and localhost are rejected).
- [Probe a URL's media metadata](https://docs.varg.ai/api-reference/files/probe-a-urls-media-metadata.md): Inspect a media URL (dimensions, duration) without storing it.
- [Check for an existing file by hash](https://docs.varg.ai/api-reference/files/check-for-an-existing-file-by-hash.md): Dedup pre-check — returns whether your account already has a file with this content hash. Lets clients skip the upload entirely.
- [Get file metadata](https://docs.varg.ai/api-reference/files/get-file-metadata.md)
- [Delete a file](https://docs.varg.ai/api-reference/files/delete-a-file.md): Soft-deletes the file (it disappears from lists; the underlying object is garbage-collected later).
- [Rate or comment on a file](https://docs.varg.ai/api-reference/files/rate-or-comment-on-a-file.md): `like`/`dislike` upserts your rating (one per account per file). `comment` is append-only.
- [List feedback for a file](https://docs.varg.ai/api-reference/files/list-feedback-for-a-file.md)
- [Remove your rating on a file](https://docs.varg.ai/api-reference/files/remove-your-rating-on-a-file.md)
- [Resolve a file to the job that created it](https://docs.varg.ai/api-reference/lineage/resolve-a-file-to-the-job-that-created-it.md): Given a `file_id` or a varg URL, returns the job that produced the file together with its recipe (prompt, model, tool, cost) — so an agent can "create something similar".
- [Estimate a price without creating a job](https://docs.varg.ai/api-reference/pricing/estimate-a-price-without-creating-a-job.md): Send the same body you would send to a generation endpoint and get the price back. Also accepts a batch format with an `items` array where each item carries `tool`, `model`, and `params`.
- [Get the price catalog (public)](https://docs.varg.ai/api-reference/pricing/get-the-price-catalog-public.md): The full model catalog with per-provider price estimates, grouped by tool. No authentication required.
- [Get balance](https://docs.varg.ai/api-reference/billing/get-balance.md): Balance breakdown in cents (credits). `available` = total − reserved. `reserved` is held by in-flight jobs and is committed on completion or released on failure.
- [Get usage records](https://docs.varg.ai/api-reference/billing/get-usage-records.md)
- [Get transaction ledger](https://docs.varg.ai/api-reference/billing/get-transaction-ledger.md): Account ledger rows (spend/topup history). Negative `amount_cents` = charge, positive = credit. Job charges carry context (`job_id`, `tool`, `model`, `billed_units`).
- [Get the authenticated account](https://docs.varg.ai/api-reference/account/get-the-authenticated-account.md)
- [List API keys](https://docs.varg.ai/api-reference/account/list-api-keys.md): Requires an app session (Supabase JWT), not an API key.
- [Create an API key](https://docs.varg.ai/api-reference/account/create-an-api-key.md): Creates a new API key. The plaintext key is returned **once** in the `api_key` field — store it securely. Requires an app session (JWT).
- [Rename an API key](https://docs.varg.ai/api-reference/account/rename-an-api-key.md)
- [Revoke an API key](https://docs.varg.ai/api-reference/account/revoke-an-api-key.md)

## OpenAPI Specs

- [openapi](https://docs.varg.ai/openapi.yaml)

## Optional

- [GitHub](https://github.com/vargHQ/varg)
- [Discord](https://discord.gg/VAecJay7R9)