# MachGen Python SDK

> Official Python client (`machgen-client`) for the MachGen generation API.

## Docs

- [Get Started](/docs): Authenticate yourself and submit your first MachGen generation task.
- [MCP Server](/docs/mcp): Connect Claude/Codex to MachGen and generate from your AI agent.
- [Available Models](/docs/models): Models you can target and the task types each one supports.
- [API Reference](/docs/reference): The request, configuration, and response models shared by the REST API and the Python client.
- [Python Client](/docs/client/python): Submit and track tasks with the official MachGen Python client.
- [Account and Billing](/docs/rest_api/account): Check your balance and what you've spent.
- [API Key IP Allowlist](/docs/rest_api/api-keys): Choose which IP addresses can use each API key.
- [Custom LoRAs](/docs/rest_api/custom-loras): Upload your own LoRA once, then use it in any generation request.
- [REST API](/docs/rest_api): Call the MachGen REST API directly over HTTPS.
- [MiniMax H3](/docs/rest_api/models/h3): Set a video's length by frame count.
- [Seedance 2.0](/docs/rest_api/models/seedance-2-0): Let the model pick the framing and length of a video.
- [Seedance 2.5](/docs/rest_api/models/seedance-2-5): Adaptive framing and length, and a 480p draft before the 1080p final.
- [Wan 3.0](/docs/rest_api/models/wan): Use a file or webpage as the reference for a video.

## Pricing & Models

- [Live pricing](/pricing.txt): current per-model API prices, plain text
- [Pricing page](/pricing): pay-as-you-go wallet pricing
- [Models](/models): image and video models with API access
- [MiniMax H3 API](/models/minimax/minimax-h3): MiniMax H3 for text, image, frame-pair, and multimodal reference video with native stereo audio.
- [MiniMax H3 Turbo API](/models/minimax/minimax-h3-turbo): MiniMax H3 Turbo for fast iterations across text, image, frame-pair, and multimodal reference video with native stereo audio.
- [LTX 2.3 Pro API](/models/lightricks/ltx-2.3-pro): LTX 2.3 Pro for cinematic 4K video, directorial control, and synchronized native audio.
- [Vidu Q3 Turbo API](/models/vidu/q3-turbo): Vidu Q3 Turbo for fast, balanced video iteration across text, image, reference, and frame-pair workflows.
- [Vidu Q3 Pro API](/models/vidu/q3): Vidu Q3 Pro for high-fidelity text, image, reference, and frame-pair video with optional synchronized audio.
- [Wan 2.2 A14B API](/models/wan-video/wan-2.2-i2v-a14b): Wan 2.2 A14B for rich visuals and detailed motion from text or stills, served on low-cost managed inference.
- [Seedance 2.5 API](/models/bytedance/seedance-2.5): Seedance 2.5 for 30-second native video with up to 50 multimodal references and directed editing.
- [Seedance 2.0 API](/models/bytedance/seedance-2.0): Seedance 2.0 for cinematic multimodal video, coherent multi-shot storytelling, and synced native audio.
- [Wan 3.0 API](/models/alibaba/wan-3.0): Wan 3.0 for production-scale video generation with 30-second shots, multimodal direction, and synchronized sound.
- [Seedance 2.0 Fast API](/models/bytedance/seedance-2.0-fast): Seedance 2.0 control and native audio tuned for faster production loops.
- [Seedance 2.0 Mini API](/models/bytedance/seedance-2.0-mini): The most efficient Seedance 2.0 tier for rapid multimodal video exploration.
- [Vidu Q3 API](/models/vidu/q3-r2v): Vidu Q3 for high-quality reference-to-video with look-locked subjects, prompt-directed camera motion, and optional audio.
- [Kling Video 3.0 API](/models/kwaivgi/kling-v3-video): Kling Video 3.0 for cinematic text-to-video and image-to-video with optional native audio.
- [Kling Video 3.0 Omni API](/models/kwaivgi/kling-o3): Compose video from prompts, reference images, and Kling elements
- [Vidu Q3 Pro Fast API](/models/vidu/q3-pro-fast): Vidu Q3 Pro Fast for low-latency, high-quality image animation with prompt-directed camera motion and optional audio.
- [Vidu S1 API](/models/vidu/s1): Vidu S1 for voice-driven realtime people, fantasy characters, and pet avatars.
- [Veo 3.1 Standard API](/models/google/veo-3.1): Google Veo 3.1 Standard for high-quality cinematic text-to-video with synchronized native audio.
- [Veo 3.1 Fast API](/models/google/veo-3.1-fast): Google Veo 3.1 Fast for cinematic native-audio video with quicker, lower-cost prompt iteration.
- [Pixverse V6 API](/models/pixverse/pixverse-v6): PixVerse V6 for cinematic, stylized video with references, prompt-directed camera motion, and optional audio.
- [Pixverse C1 API](/models/pixverse/pixverse-c1): PixVerse C1 for cinematic action, VFX-heavy scenes, and reference-guided visual continuity.
- [HappyHorse 1.1 API](/models/alibaba/happyhorse-1.1): HappyHorse 1.1 for high-quality general video and stronger multi-reference subject consistency.
- [HappyHorse 1.0 API](/models/alibaba/happyhorse-1.0): HappyHorse 1.0 for balanced-cost, general-purpose video from text, a first frame, or references.
- [Grok Imagine Video API](/models/xai/grok-imagine-video): Grok Imagine Video for fast, expressive text-to-video concepts and social-ready visual exploration.
- [Grok Imagine Video 1.5 API](/models/xai/grok-imagine-video-1.5): Grok Imagine Video 1.5 for realistic still-image motion, object interaction, and synchronized generated audio.
- [Topaz Precision API](/models/topaz/video-precision): Topaz Precision for detail-preserving video upscaling to 1080p, 2K, 4K, or 8K.
- [Topaz Precision Animation API](/models/topaz/video-precision-animation): Topaz Precision Animation for upscaling hand-drawn and cel animation.
- [Topaz Generative API](/models/topaz/video-generative): Topaz Generative for diffusion-based video restoration up to 4K.
- [Topaz Generative Fast API](/models/topaz/video-generative-fast): Topaz Generative Fast for cheaper generative video restoration on longer clips.
- [FLUX.2 Dev API](/models/black-forest-labs/flux-2-dev): FLUX.2 Dev for high-quality open-weight image generation, multi-image editing, and fast experimentation.
- [HiDream O1 API](/models/hidream/hidream-o1): HiDream O1 for high-detail text-to-image across fantasy, illustration, and photoreal visual styles.
- [Grok Imagine Image API](/models/xai/grok-imagine-image): Grok Imagine Image for fast, low-cost text-to-image with a bold, expressive visual style.
- [Grok Imagine Image Quality API](/models/xai/grok-imagine-image-quality): Grok Imagine Image Quality for higher-fidelity generation, cleaner detail, and reference-based editing.
- [Nano Banana 2 API](/models/google/nano-banana-2): Nano Banana 2 for fast, high-fidelity Gemini image generation, natural-language edits, and clear typography.
- [Nano Banana 2.1 API](/models/google/nano-banana-2.1): Nano Banana 2.1 for detailed image generation, readable text, and reference-guided edits.
- [Nano Banana Pro API](/models/google/nano-banana-pro): Nano Banana Pro for high-quality Gemini image generation, complex briefs, typography, and controlled edits.
- [Seedream 5.0 Lite API](/models/bytedance/seedream-5.0-lite): Seedream 5.0 Lite for efficient high-resolution image creation, layout-aware prompts, and reference edits.
- [GPT Image 2 API](/models/openai/gpt-image-2): GPT Image 2 for high-fidelity image generation, precise instruction following, typography, and revisions.
- [GPT Image 2.5 Sunburst API](/models/openai/gpt-image-2.5-sunburst): GPT Image 2.5 Sunburst for detailed image generation and precise edits.
- [GPT Image 2.5 Flare API](/models/openai/gpt-image-2.5-flare): GPT Image 2.5 Flare for fast image generation and editing.
- [Topaz Image Precision API](/models/topaz/image-precision): Topaz Image Precision for faithful 2x or 4x image upscaling.
- [Topaz Image Generative API](/models/topaz/image-generative): Topaz Image Generative for prompt-steerable image upscaling.
- [YuE2 API](/models/machgen/yue2): YuE2 for vocal songs from your lyrics and instrumental tracks from a style description.
- [ElevenLabs v3 API](/models/elevenlabs/eleven-v3): ElevenLabs v3 for expressive speech and multi-speaker dialogue generated in one pass.
- [ElevenLabs Sound Effects v2 API](/models/elevenlabs/eleven-sfx-v2): ElevenLabs SFX v2 for foley and designed sound effects written as a description.
- [ElevenLabs Music v2 API](/models/elevenlabs/eleven-music-v2): ElevenLabs Music v2 for instrumental and vocal tracks, from a prompt or a section plan.

## Full text

- [llms-full.txt](/llms-full.txt): every page as Markdown in one file