# Meshy v7 Multi-Image-to 3D — Atlas Cloud API > Meshy v7 Multi-Image-to-3D reconstructs one textured 3D model from up to four views of the same object, so the sides and back follow your reference instead of being inferred. Supports independent multi-view texture references and PBR maps up to 8K. This is the machine-readable API reference for **Meshy v7 Multi-Image-to 3D** on Atlas Cloud, a unified API platform for 400+ AI models across text, image, video, audio and 3D. - **Model ID**: `meshy-v7/multi-image-to-3d` - **Built by**: MESHY - **Modality**: Image - **Model page**: https://www.atlascloud.ai/models/meshy-v7/multi-image-to-3d - **API key**: https://www.atlascloud.ai/console/api-keys - **Docs**: https://www.atlascloud.ai/docs ## Pricing on Atlas Cloud - $0.02 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: "meshy-v7/multi-image-to-3d"`. 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 meshy-v7/multi-image-to-3d -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**: `meshy-v7/multi-image-to-3d` ## 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: `"meshy-v7/multi-image-to-3d"` - Options: "meshy-v7/multi-image-to-3d" - **`reference_images`** (`array[string]`, _required_): 1 to 4 images of the same object from different angles. Each may be a public URL, base64 data URI, or asset reference (asset://); formats jpg/jpeg/png. The first image is the primary (front) view — the order of the rest does not matter. - Min items: 1 - Max items: 4 - **`ultra_mode`** (`boolean`, _optional_): Higher-fidelity geometry with finer surface detail. Takes longer and costs 5 extra credits. - Default: `false` - **`should_texture`** (`boolean`, _optional_): Generate textures. Turn it off for an untextured mesh — faster and cheaper. - Default: `true` - **`enable_pbr`** (`boolean`, _optional_): Also generate PBR maps (metallic, roughness, normal) alongside the base colour. Requires `should_texture`. - Default: `false` - **`texture_resolution`** (`string`, _optional_): Base colour texture resolution: 2k (2048²), 4k (4096²) or 8k (8192²). 8k costs 5 extra credits. - Default: `"2k"` - Options: "2k", "4k", "8k" - **`texture_prompt`** (`string`, _optional_): Extra prompt describing the surface material. Up to 800 characters. Cannot be combined with `texture_images`. - **`texture_image`** (`string`, _optional_): Single reference image guiding the texture, as a public URL, base64 data URI, or asset reference (asset://). Cannot be combined with `texture_images`. `texture_prompt` takes priority when both are given. - **`texture_images`** (`array[string]`, _optional_): 1 to 4 multi-view images guiding the texture only, independent of `reference_images` (different images and a different count are fine). The first is the primary (front) view. Cannot be combined with `texture_image` or `texture_prompt`. - Min items: 1 - Max items: 4 - **`image_enhancement`** (`boolean`, _optional_): Pre-process the source images for a better reconstruction. Turn it off to preserve their exact appearance with no style processing. - Default: `true` - **`remove_lighting`** (`boolean`, _optional_): Strip baked highlights and shadows from the base colour texture so the model sits correctly under your own lighting. Supported on Meshy 7 for multi-image input. - Default: `true` - **`should_remesh`** (`boolean`, _optional_): Run the remesh phase. Leave off for the highest-quality mesh; turn it on when you need a specific polygon budget via `target_polycount` or `decimation_mode`. - Default: `false` - **`topology`** (`string`, _optional_): Topology of the remeshed output. `quad` produces a quad-dominant mesh, `triangle` a decimated triangle mesh. Requires `should_remesh`. - Default: `"triangle"` - Options: "triangle", "quad" - **`target_polycount`** (`integer`, _optional_): Target face count for the remesh phase: 100-300000, default 30000. Requires `should_remesh`, and is ignored when `decimation_mode` is set. - Default: `30000` - Min: 100 - Max: 300000 - **`decimation_mode`** (`integer`, _optional_): Adaptive decimation level, used instead of an exact `target_polycount`. 1 = ultra, 2 = high, 3 = medium, 4 = low polycount. Requires `should_remesh`, and takes priority over `target_polycount`. - Options: 1, 2, 3, 4 - **`save_pre_remeshed_model`** (`boolean`, _optional_): Also return the GLB as it was before the remesh phase, so you can compare the two. Requires `should_remesh`. - Default: `false` - **`pose_mode`** (`string`, _optional_): Force a rest pose for character meshes. Empty applies no specific pose. - Default: `""` - Options: "", "a-pose", "t-pose" - **`target_formats`** (`array[string]`, _optional_): Which 3D file formats to return. Defaults to GLB alone; ask for more only when you need them, since each extra format adds generation and transfer time. - Default: `["glb"]` - Min items: 1 - Max items: 6 - **`auto_size`** (`boolean`, _optional_): Estimate the object's real-world height with AI vision and scale the model accordingly. - Default: `false` - **`origin_at`** (`string`, _optional_): Where the model origin sits when `auto_size` is on. - Default: `"bottom"` - Options: "bottom", "center" - **`alpha_thumbnail`** (`boolean`, _optional_): Also render a transparent-background (RGBA) preview image alongside the normal thumbnail. - Default: `false` - **`multi_view_thumbnails`** (`boolean`, _optional_): Also render right / back / left preview images so the model can be reviewed from four angles without downloading the mesh. Adds roughly 3 seconds. - Default: `false` - **`moderation`** (`boolean`, _optional_): Screen every source image and the texture prompt for harmful content before generating. A flagged request does not proceed to generation. - Default: `false` **Required Parameters Example**: ```json { "model": "meshy-v7/multi-image-to-3d", "reference_images": [ "" ] } ``` **Full Example**: ```json { "model": "meshy-v7/multi-image-to-3d", "reference_images": [ "" ], "ultra_mode": false, "should_texture": true, "enable_pbr": false, "texture_resolution": "2k", "texture_prompt": "", "texture_image": "", "texture_images": [ "" ], "image_enhancement": true, "remove_lighting": true, "should_remesh": false, "topology": "triangle", "target_polycount": 30000, "decimation_mode": 0, "save_pre_remeshed_model": false, "pose_mode": "", "target_formats": [ "glb" ], "auto_size": false, "origin_at": "bottom", "alpha_thumbnail": false, "multi_view_thumbnails": false, "moderation": false } ``` ### Output Schema The API returns the following output format: - **`code`** (`integer`, _optional_): - **`message`** (`string`, _optional_): - **`data`** (`string`, _optional_): **Example Response**: ```json { "code": 0, "message": "", "data": null } ``` ## 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": "meshy-v7/multi-image-to-3d", "reference_images": [ "" ], "ultra_mode": false, "should_texture": true, "enable_pbr": false, "texture_resolution": "2k", "texture_prompt": "", "texture_image": "", "texture_images": [ "" ], "image_enhancement": true, "remove_lighting": true, "should_remesh": false, "topology": "triangle", "target_polycount": 30000, "decimation_mode": 0, "save_pre_remeshed_model": false, "pose_mode": "", "target_formats": [ "glb" ], "auto_size": false, "origin_at": "bottom", "alpha_thumbnail": false, "multi_view_thumbnails": false, "moderation": 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/meshy-v7/multi-image-to-3d) ## About this model _Vendor-supplied description. Any pricing or endpoint mentioned below refers to_ _other platforms — use the Atlas Cloud values above._ ### Meshy 7 Multi-Image-to-3D **Meshy 7 Multi-Image-to-3D** reconstructs one textured 3D model from up to four photographs of the same object. Where a single image forces the model to invent the sides and back, several views let it read them — so the parts of an object that a camera can see are the parts that come out right. #### Why Choose This? - **The back is no longer a guess** — feed front, side and rear views and the reconstruction follows your reference rather than a prior. - **Texture references of their own** — `texture_images` guides the surface independently of the geometry inputs. - **One call, one model** — geometry and PBR textures in the same task. - **Polygon budgets you control** — an exact face count, an adaptive decimation level, or the raw high-detail mesh. - **Review before you download** — optional four-angle preview renders and a transparent-background thumbnail. #### Parameters | Parameter | Required | Description | | --- | --- | --- | | `model` | Yes | `meshy-v7/multi-image-to-3d` | | `reference_images` | Yes | 1–4 images of the same object from different angles. Each: public URL, base64 data URI, or `asset://`. jpg/jpeg/png. **The first image is the front view.** | | `ultra_mode` | No | Finer surface detail. Slower, +5 credits. Default: `false` | | `should_texture` | No | Generate textures. Default: `true`; `false` returns a bare mesh for less. | | `enable_pbr` | No | Also produce metallic / roughness / normal maps. Default: `false` | | `texture_resolution` | No | `2k` (default), `4k`, or `8k`. | | `texture_prompt` | No | Describe the surface material, up to 800 characters. | | `texture_image` | No | A single reference image for the texture. | | `texture_images` | No | 1–4 multi-view texture references. Cannot be combined with `texture_image` or `texture_prompt`. | | `image_enhancement` | No | Pre-process the inputs for a better reconstruction. Default: `true` | | `remove_lighting` | No | Strip baked highlights and shadows. S _(Description truncated. Full text on the model page.)_ --- Atlas Cloud — one API for 400+ AI models. Model page: https://www.atlascloud.ai/models/meshy-v7/multi-image-to-3d · Docs: https://www.atlascloud.ai/docs