MOSS-VL

BetaMultimodal understanding

MOSS-VL accepts text instructions with images, video, or multiple images for image understanding, video understanding, OCR, document parsing, visual question answering, and cross-media content analysis. Each request supports up to 5 images or 1 video; images and video cannot be mixed.

View API reference

ENDPOINT

POST /v1/responses

Current request path

Input

Text + images / video

Content accepted by the request

Output

Text

Content returned on success

Model overview

Multimodal understanding model. Use the stable model ID moss-vl-1.0 by default; to reproduce a fixed version, pass moss-vl-1.0-2026-07-08 in model.

Billing

Free for a limited time.

0 credits View full pricing

Modality support

TTextInput and output

AAudioNot supported

IImageInput only

VVideoInput only

Supported endpoints

Single-speaker speech/v1/audio/speechNot supported

Multi-speaker dialogue speech/v1/audio/speech/speakersNot supported

Audio transcription/v1/audio/transcriptionsNot supported

Voice design/v1/audio/voice/generationsNot supported

Multimodal understanding/v1/responsesSupported

Realtime video understanding/v1/realtimeNot supported

Image generation/v1/images/generationsNot supported

Capabilities

PerformanceHigh

Inference speedMedium

Streaming Not supported

Async Not supported

Image description, OCR, and document content understanding
Video summaries, event understanding, and question answering
Up to 5 images or 1 video per request
Supports file ID and URL input

Versions and snapshots

Model IDmoss-vl-1.0

Snapshotmoss-vl-1.0-2026-07-08Latest