Console

ModelsMOSS-Image-SFT-1.0

MOSS-Image-SFT-1.0

Image generation

MOSS-Image-SFT-1.0 is the SFT (supervised fine-tuned) version of MOSS-Image-Base-1.0. It uses the same image generation endpoint and parameters, supports long text and text rendering, and can enhance the prompt and generate the image within the same request.

ENDPOINT

POST /v1/images/generations

Current request path

Input

Text

Content accepted by the request

Output

Image

Content returned on success

Model overview

Text-to-image model, the SFT version of MOSS-Image-Base-1.0. Use the stable model ID moss-image-sft-1.0 by default; to reproduce a fixed version, pass moss-image-sft-1.0-2026-10-09 in model.

Billing

Billing methodCNY price 40% OFFActual deduction 40% OFF
Low · About 1 MPTotal pixels ≤ 1,048,576CNY 0.20CNY 0.12 / image42.40 credits / image
High · About 2 MPTotal pixels above 1,048,576, up to 2,097,152CNY 0.30CNY 0.18 / image63.60 credits / image
Prompt enhancementBilled additionally when enabledCNY 0.05CNY 0.03 / request10.60 credit / request

Images are billed in tiers by the total pixels of each image (width × height). With prompt_enhance: true, prompt enhancement is billed separately per request.
For example, generating one 2K image with one prompt enhancement costs 4.20 credits (CNY 0.21).
Each request is billed by actual usage. CNY prices are for reference; the final charge follows the credits bill. View full pricing

Modality support

TTextInput only
AAudioNot supported
IImageOutput only
VVideoNot supported

Supported endpoints

Chat Completions /v1/chat/completionsNot supported
Responses /v1/responsesNot supported
Messages /v1/messagesNot supported
Count input tokens /v1/messages/count_tokensNot supported
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
Realtime interaction /v1/realtimeNot supported
Image generation /v1/images/generationsSupported

The built-in Image prompt enhancement uses an enhancer model version selected by the server; the client does not specify it. Off by default; enable it with prompt_enhance: true.

Capabilities

PerformanceHigh
Inference speedMedium
Streaming Not supported
Async task API Not supported
  • Supports long text up to 4,000 characters
  • Supports text rendering
  • Supports negative prompts
  • Supports layout generation

Rate limits

LimitCapScope
Requests per minute (RPM)70 requestsShared across all model endpoints except MOSS-Live-1.0 and Nex-N2.6 for the same user
Requests per day (RPD)30,000 requestsShared across all model endpoints except MOSS-Live-1.0 and Nex-N2.6 for the same user
Concurrency3Each user can process up to 3 generation tasks at the same time on this model.

View full request limits

Versions and snapshots

Model IDmoss-image-sft-1.0
Snapshotmoss-image-sft-1.0-2026-10-09Latest