Meshy 7.1 Text to 3D
meshy/v7.1/text-to-3dGenerate 3D models from text with Meshy 7.1 Text to 3D from fal through one Spiral API key. Meshy 7.1 is the latest standard model from Meshy. It generates 3D models from text prompts, with untextured preview and textured full modes. It runs in the background and returns 3D models and images. Pay as you go, with no subscription.
Pricing
mode: preview $1.04, otherwise $1.56 per model. enable_rigging adds $0.26 per model. enable_rigging and enable_animation add $0.156 per model.
Limits
- Runs
- In the background
- Returns
- 3d, image
- Longest queue wait
- 1 hour
- Run deadline
- 30 minutes
- Longest run, end to end
- 225 minutes
- Largest inline input
- 1 MB
- Largest input file
- 2 GB
- Largest output
- 5 GB
Input
| Field | Kind | Default | Description |
|---|---|---|---|
prompt required | text | — | Describe what kind of object the 3D model is. Maximum 600 characters. |
mode | one of preview, full | full | Generation mode. 'preview' returns untextured geometry only, 'full' returns textured model (preview + refine). |
seed | number (whole) | — | Seed for reproducible results. Same prompt and seed usually generate the same result. |
model_type | one of standard | standard | Type of 3D mesh generation. Only 'standard' is supported; Smart Topology has its own endpoints (/t2.5 and /t2). |
topology | one of quad, triangle | triangle | Specify the topology of the generated model. Quad for smooth surfaces, Triangle for detailed geometry. |
target_polycount | number (whole) 100 to 300000 | 30000 | Target number of polygons in the generated model. |
should_remesh | true or false | true | Whether to enable the remesh phase. When false, returns unprocessed triangular mesh. |
symmetry_mode | one of off, auto, on | auto | Controls symmetry behavior during model generation. |
enable_pbr | true or false | false | Generate PBR Maps (metallic, roughness, normal) in addition to base color. |
is_a_t_pose | true or false | false | Deprecated: use pose_mode instead. When true, generates a T-pose model. |
pose_mode | one of a-pose, t-pose, | Pose mode for the generated model. 'a-pose' generates an A-pose, 't-pose' generates a T-pose, empty string for no specific pose. | |
enable_prompt_expansion | true or false | false | Whether to enable prompt expansion. This will use a large language model to expand the prompt with additional details while maintaining the original meaning. |
texture_prompt | text | — | Additional text prompt to guide the texturing process (only used in 'full' mode). Maximum 800 characters. |
texture_image_url | URL or file | — | 2D image to guide the texturing process (only used in 'full' mode) |
texture_model | text | — | Model for the texturing (refine) step, only used in 'full' mode. Leave unset to texture with the endpoint's model. meshy-n1 makes a 2K base color texture from a reference image only: it rejects texture_prompt, enable_pbr, and quad topology with remeshing, and needs texture_image_url unless the endpoint is v7.1 (whose preview stores a reference image). |
enable_rigging | true or false | false | Automatically rig the generated model as a humanoid character. Includes basic walking and running animations. Best results with humanoid characters that have clearly defined limbs. |
rigging_height_meters | number | 1.7 | Approximate height of the character in meters. Only used when enable_rigging is true. |
skeleton_template | one of mixamo, ue5_mannequin, unity_humanoid, maya_human_ik, blender_rigify_basic_human, max_biped, cc5_hd_neutral_base, daz_genesis9, meshyrig | — | Skeleton template for the rig. Animations made from the rig use the same skeleton. Leave unset for Meshy's default skeleton (meshyrig). Some models can only be rigged on meshyrig without fingers; those requests fail with a 422. Only used when enable_rigging is true. |
include_fingers | true or false | false | Rig the character's fingers. Works with any skeleton template. Only used when enable_rigging is true. |
enable_animation | true or false | false | Apply an animation preset to the rigged model. Requires enable_rigging to be true. |
animation_action_id | number (whole) | 92 | Animation preset ID from Meshy's library. Only used when enable_animation is true; in that case it must be in [0, 696] (` |
geometry_resolution | one of standard, 2k, 4k | — | Geometry resolution. 4k is only available with Meshy-7.1. |
ultra_mode | true or false | false | Deprecated: use geometry_resolution='2k'. If both fields are provided, geometry_resolution takes precedence. |
Same request in code
curl https://api.spiral.ai/v1/runs \
-H "Authorization: Bearer $SPIRAL_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $(uuidgen)" \
-d '{
"provider": "fal",
"model": "meshy/v7.1/text-to-3d",
"input": {
"prompt": "A rustic, antique wooden treasure chest with a curved, domed lid, constructed from weathered, dark brown planks exhibiting prominent wood grain and subtle…"
}
}'Frequently asked questions
How much does Meshy 7.1 Text to 3D cost?
`mode`: preview $1.04, otherwise $1.56 per model. `enable_rigging` adds $0.26 per model. `enable_rigging` and `enable_animation` add $0.156 per model. You pay as you go from one Spiral balance, with no subscription, and a failed run is not charged.
How do I use Meshy 7.1 Text to 3D through an API?
Send `POST /v1/runs` with `"provider": "fal"`, `"model": "meshy/v7.1/text-to-3d"` and the input fields on this page, using one Spiral API key. It runs in the background: poll the run, or name a webhook to be told when it finishes.
Do I need an account with fal to use Meshy 7.1 Text to 3D?
No. Spiral runs it on its own fal account and charges your Spiral balance.
Can my AI agent use Meshy 7.1 Text to 3D?
Yes. Agents find and run it through Spiral's MCP server, CLI or agent skill, with the same API key.