fov_xnumber
Known horizontal field of view in degrees. Estimated from the image when omitted.
Minimum: 1
Maximum: 179
MoGe-3
microsoft/moge-3MoGe-3 by Microsoft estimates metric 3D geometry from a single image, refining depth with a sparse 3D refiner, and returns a textured GLB mesh, a PLY point cloud, and depth, normal and mask maps.
Try the model
Playground
Input
PNG, JPEG, WebP, or GIF · 20 MiB maximum
Input RGB image to reconstruct.
This field is required.
Advanced settings
05
Number of sparse 3D refinement passes over the predicted depth. 0 skips the refiner. Range: 0 to 5.
09
Processing resolution level (0-9). Higher values give finer detail but run slower; output size always matches the input. Range: 0 to 9.
Known horizontal field of view in degrees. Estimated from the image when omitted. Range: 1 to 179.
Mask out invalid regions (sky, background, uncertain) in the reconstruction.
ResultIdle
Your output will appear here
Complete the inputs, then click Run.
MoGe-3
MoGe-3 by Microsoft estimates metric 3D geometry from a single image, refining depth with a sparse 3D refiner, and returns a textured GLB mesh, a PLY point cloud, and depth, normal and mask maps.
Fields
| Field | Type | Default | Notes |
|---|---|---|---|
image | image URL, required | — | Input RGB image to reconstruct. |
num_refinement_steps | integer 0–5 | 3 | Number of sparse 3D refinement passes over the predicted depth. 0 skips the refiner. |
resolution_level | integer 0–9 | 9 | Processing resolution level (0-9). Higher values give finer detail but run slower; output size always matches the input. |
fov_x | number 1–179 | — | Known horizontal field of view in degrees. Estimated from the image when omitted. |
apply_mask | boolean | true | Mask out invalid regions (sky, background, uncertain) in the reconstruction. |
Output
url: Main 3D file (GLB unless another export format is selected).point_cloud: Additional file returned by the provider.depth_map: Additional file returned by the provider.normal_map: Additional file returned by the provider.mask: Additional file returned by the provider.
Billing
$0.075 per request. Failed runs are not charged.
Also on SandBase
Other Microsoft 3D endpoints: microsoft/moge-2.
Start building
Send your first request
Submit this model through the SandBase Unified Run API.
# The quoted heredoc keeps Unicode and shell metacharacters unchanged.
result=$(curl --fail-with-body --silent \
-X POST "https://api.sandbase.ai/v1/run" \
-H "Authorization: Bearer $SANDBASE_API_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'SANDBASE_JSON'
{
"model": "microsoft/moge-3",
"image": "https://storage.googleapis.com/falserverless/example_inputs/dog.png",
"apply_mask": true,
"resolution_level": 9,
"num_refinement_steps": 3
}
SANDBASE_JSON
)
run_id=$(printf '%s' "$result" | jq -r .id)
for attempt in $(seq 1 120); do
status=$(printf '%s' "$result" | jq -r .status)
case "$status" in completed|failed|timeout) break ;; esac
sleep 2
result=$(curl --fail-with-body --silent \
-H "Authorization: Bearer $SANDBASE_API_KEY" \
"https://api.sandbase.ai/v1/run/$run_id")
done
status=$(printf '%s' "$result" | jq -r .status)
[ "$status" = completed ] || { echo "Generation ended: $status" >&2; exit 1; }
printf '%s\n' "$result"2. Authentication
Create a SandBase API key and set it in your server environment. All requests use Bearer authentication.
export SANDBASE_API_KEY="YOUR_API_KEY"Keep the key on your server. Do not expose it in browser code or commit it to a repository.
3. Queue
Submit a request
POST to /v1/run with this model ID and its inputs. The request example above includes submission and polling.
Fetch request status and result
Use the returned run ID to poll until the status is completed, failed or timeout. Read output only after completed.
curl "https://api.sandbase.ai/v1/run/$RUN_ID" \
-H "Authorization: Bearer $SANDBASE_API_KEY"4. Files
Hosted files
Pass a publicly accessible file URL in the model’s image, video or audio field. Supported formats and limits are shown beside each Playground input.
Uploading images
Use the upload or replace control on supported image fields in the Playground. Video URL fields accept a hosted driving video; they do not imply video-upload support.
5. Schema
Input
imagestringrequired
Input RGB image to reconstruct.
apply_maskboolean
Mask out invalid regions (sky, background, uncertain) in the reconstruction.
Default: true
resolution_levelinteger
Processing resolution level (0-9). Higher values give finer detail but run slower; output size always matches the input.
Default: 9
Minimum: 0
Maximum: 9
num_refinement_stepsinteger
Number of sparse 3D refinement passes over the predicted depth. 0 skips the refiner.
Default: 3
Minimum: 0
Maximum: 5
{
"model": "microsoft/moge-3",
"image": "https://storage.googleapis.com/falserverless/example_inputs/dog.png",
"apply_mask": true,
"resolution_level": 9,
"num_refinement_steps": 3
}Output
A completed run exposes its result in output. The following is the model’s saved example, not a new generation.
No example output has been published yet. Use Playground → Result → JSON after a completed run to inspect your response.