# MiniMax H3 Image Highspeed Edit — Atlas Cloud API > The Highspeed serving mode of MiniMax's open-weight H3-Image model, editing and composing images from 1-16 reference images and a natural-language instruction at 3-4x the speed of mainstream alternatives for latency-sensitive workflows. This is the machine-readable API reference for **MiniMax H3 Image Highspeed Edit** on Atlas Cloud, a unified API platform for 400+ AI models across text, image, video, audio and 3D. - **Model ID**: `minimax/h3-image-highspeed/edit` - **Built by**: MiniMax - **Modality**: Image - **Model page**: https://www.atlascloud.ai/models/minimax/h3-image-highspeed/edit - **API key**: https://www.atlascloud.ai/console/api-keys - **Docs**: https://www.atlascloud.ai/docs ## Pricing on Atlas Cloud - $0.046 per image - Pay-as-you-go. No minimum spend, no subscription required. > **These are the authoritative Atlas Cloud rates for this model.** Any price that > appears in the vendor description further down refers to a different platform or > a different model variant and does not apply here. ## Use this model from an AI agent Atlas Cloud ships three first-party integration surfaces. All three authenticate with the same API key via the `ATLASCLOUD_API_KEY` environment variable. ### MCP server The official MCP server (`atlascloud-mcp`) exposes this model to any MCP-compatible host — Claude Code, OpenAI Codex, Cursor, Gemini CLI, Goose, Claude Desktop. One-line install: ```bash # Claude Code claude mcp add atlascloud -- npx -y atlascloud-mcp # OpenAI Codex CLI codex mcp add atlascloud -- npx -y atlascloud-mcp # Gemini CLI gemini mcp add atlascloud -- npx -y atlascloud-mcp export ATLASCLOUD_API_KEY="your-api-key" ``` Then ask in plain English; the agent calls `atlas_generate_image` with `model: "minimax/h3-image-highspeed/edit"`. The server fetches each model's schema and validates parameters before submitting, so invalid requests fail fast without spending credits. MCP docs: https://www.atlascloud.ai/docs/mcp-server ### Agent Skills `atlas-cloud-skills` is a portable skill package (API reference, code templates in Python / Node.js / cURL, model IDs with pricing) for Claude Code, Cursor, Codex and 12+ other agents: ```bash npx skills add AtlasCloudAI/atlas-cloud-skills export ATLASCLOUD_API_KEY="your-api-key" ``` Skills docs: https://www.atlascloud.ai/docs/skills ### CLI The `atlas` binary runs Atlas Cloud from a terminal or CI script. Async media jobs are polled and downloaded automatically (use `--no-download` when a script only needs the output URLs): ```bash # Install (Homebrew, npm, or shell installer) brew install AtlasCloudAI/tap/atlascloud # npm install -g atlascloud-cli # curl -fsSL https://raw.githubusercontent.com/AtlasCloudAI/cli/main/install.sh | sh atlas auth login atlas generate image minimax/h3-image-highspeed/edit -p "Your prompt here" ``` CLI docs: https://www.atlascloud.ai/docs/cli ## HTTP API reference - **Submit endpoint (POST)**: `https://api.atlascloud.ai/api/v1/model/generateImage` — start an async generation; returns a `prediction_id` - **Poll endpoint (GET)**: `https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}` — poll this until the prediction finishes - **Model ID**: `minimax/h3-image-highspeed/edit` ## API Information This model can be used via our HTTP API or more conveniently via our client libraries. See the input and output schema below, as well as the usage examples. ### Input Schema The API accepts the following input parameters: - **`model`** (`string`, _required_): model name - Default: `"minimax/h3-image-highspeed/edit"` - **`prompt`** (`string`, _required_): Text description of the edit, up to 32,000 characters. Refer to the input images by their order in reference_images (image 1, image 2, ...) and give each one a role, e.g. "Keep the product from image 1, use the colors of image 2 and the layout of image 3." State what to change and what must stay the same. Prompt expansion is always applied to edits. - **`reference_images`** (`array[string]`, _required_): The images to edit, also used as references for subjects, products, people, style or layout. Accepts 1 to 16 images; their order sets the image numbers used in the prompt (the first is image 1). Supported formats: JPG, JPEG, PNG, WEBP, HEIC, HEIF. Each image must be at most 64MB, with width and height within [128, 8192] px and an aspect ratio (long side / short side) of at most 4:1. Supports both a public URL and a base64-encoded image (e.g., data:image/png;base64,) for each item. - Min items: 1 - Max items: 16 - **`size`** (`string`, _optional_): Output image size in width*height format (e.g., 2048*2048, 2720*1536, 4096*4096). The area (width × height) must be between 655,360 and 20,000,000 pixels and the aspect ratio (width:height) between 1:3 and 3:1, bounds inclusive, which limits each side to 468 to 7745 pixels; the size is used exactly as given, never scaled or rounded. Common sizes at the 2K / 4K tiers: 1:1 2048*2048 / 4096*4096; 4:3 2368*1760 / 4736*3552; 3:4 1760*2368 / 3552*4736; 16:9 2720*1536 / 5472*3072; 9:16 1536*2720 / 3072*5472; 3:1 3552*1184 / 7104*2368; 1:3 1184*3552 / 2368*7104. The special value "auto" (default) lets the model choose an aspect ratio based on the reference images and the prompt, at the 4K tier. - Default: `"auto"` - Min: 468 - Max: 7745 - **`num_images`** (`integer`, _optional_): The number of images to generate, from 1 to 10. All images share the same prompt and settings, giving different variations of it. - Default: `1` - Min: 1 - Max: 10 - **`output_format`** (`string`, _optional_): Output image file format. Transparent backgrounds are not supported. - Default: `"jpeg"` - Options: "jpeg", "png", "webp" - **`output_compression`** (`integer`, _optional_): Compression quality for jpeg and webp output, from 0 to 100; higher values give better quality and larger files. Ignored for png. - Default: `95` - Min: 0 - Max: 100 - **`enable_sync_mode`** (`boolean`, _optional_): If set to true, the function will wait for the result to be generated and uploaded before returning the response. It allows you to get the result directly in the response. This property is only available through the API. - Default: `false` - **`enable_base64_output`** (`boolean`, _optional_): If enabled, the output will be encoded into a BASE64 string instead of a URL. This property is only available through the API. - Default: `false` **Required Parameters Example**: ```json { "model": "minimax/h3-image-highspeed/edit", "prompt": "", "reference_images": [ "" ] } ``` **Full Example**: ```json { "model": "minimax/h3-image-highspeed/edit", "prompt": "", "reference_images": [ "" ], "size": "auto", "num_images": 1, "output_format": "jpeg", "output_compression": 95, "enable_sync_mode": false, "enable_base64_output": false } ``` ### Output Schema The API returns the following output format: - **`code`** (`integer`, _optional_): Status code of the call; 200 on success. - **`message`** (`string`, _optional_): Status message of the call. - **`data`** (`object`, _optional_): The prediction: id, model, status, outputs, urls and created_at. Poll GET /api/v1/model/prediction/{request_id} with data.id until status is completed or failed. **Example Response**: ```json { "code": 0, "message": "", "data": {} } ``` ## Usage Examples ### cURL ```bash # Step 1: Start generation (async) curl -X POST "https://api.atlascloud.ai/api/v1/model/generateImage" \ -H "Authorization: Bearer $ATLASCLOUD_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "minimax/h3-image-highspeed/edit", "prompt": "", "reference_images": [ "" ], "size": "auto", "num_images": 1, "output_format": "jpeg", "output_compression": 95, "enable_sync_mode": false, "enable_base64_output": false }' # Response will contain: {"code": 200, "data": {"id": "prediction_id", "status": "processing"}} # Step 2: Poll for result (replace {prediction_id} with the id returned above) curl -X GET "https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}" \ -H "Authorization: Bearer $ATLASCLOUD_API_KEY" # Keep polling until status is "completed", "succeeded" or "failed" # When completed, outputs will contain the generated content URL(s) ``` ## Additional Resources ### Documentation - [Model Playground](https://www.atlascloud.ai/models/minimax/h3-image-highspeed/edit) ## About this model _Vendor-supplied description. Any pricing or endpoint mentioned below refers to_ _other platforms — use the Atlas Cloud values above._ ### MiniMax H3-Image Highspeed Edit **Developer:** MiniMax **Model ID:** `minimax/h3-image-highspeed/edit` **Underlying Model:** MiniMax-H3-Image-Highspeed **Release Date:** November 2026 #### Overview MiniMax H3-Image is MiniMax's open-weight image generation and editing model, built on the MiniMax H3 foundation model and released in November 2026. A single checkpoint handles text-to-image, reference-guided generation and instruction-based editing. MiniMax positions it as state of the art among open-weight image models, with image quality in the same class as Nano Banana 2.1, stronger editing, and the strongest instruction following of any open-source image model. It renders natively at about 16 megapixels, so a 4K (4096×4096) image comes straight out of the model without an upscaling pass. MiniMax reports a 4K image in under 10 seconds and a 2K edit in roughly 10–15 seconds, about three times faster than Nano Banana 2.1. This endpoint, `minimax/h3-image-highspeed/edit`, is the **image editing** path of **MiniMax-H3-Image-Highspeed**: you supply 1 to 16 reference images plus a natural-language instruction, and the model returns 1 to 10 edited or newly composed images. References can be the image being edited or sources of people, products, props, materials, color and layout, all combined into one coherent picture. Prompt expansion is always applied, so short instructions work well. **Highspeed** is the faster serving mode of the same model. It accepts exactly the same parameters and limits as the Standard model and reaches **3–4× the speed of mainstream alternatives**, compared with 1.5–2× for Standard, at a higher image-token rate. Use it when latency matters more than unit cost. AtlasCloud exposes the model family as four endpoints. Two underlying models, **MiniMax-H3-Image** (Standard) and **MiniMax-H3-Image-Highspeed**, are each split by input type into a text-to-image endpoint and an edit endpoint: _(Description truncated. Full text on the model page.)_ --- Atlas Cloud — one API for 400+ AI models. Model page: https://www.atlascloud.ai/models/minimax/h3-image-highspeed/edit · Docs: https://www.atlascloud.ai/docs