Featured models from the catalog

Streaming Gemini speech for low-latency playback
Fine-grained style and expression controls

8K ctxaudio outstreaming

Lowest rate

$0.90 / $18.00

per 1M tokens

Billing

Metered

see pricing

Alibaba (Qwen)

Qwen3 TTS Flash

Content-oriented TTS with multilingual preset voices
Natural delivery for standard narration

audio outstreaming

Task rate

$0.08

Billing

Per task

see pricing

MiniMax

MiniMax H3

Multimodal Hailuo video generation with reference media
720P through 4K short-form output with image, video, audio, and last-frame inputs

multimodalvideo output

Task rate

$0.08

per second

Billing

Per task

see pricing

Alibaba (Qwen)

Qwen3 TTS Flash Realtime

Low-latency bidirectional streaming TTS
Preset multilingual voices for immediate playback

audio outstreaming

Task rate

$0.104

Billing

Per task

see pricing

Alibaba (Qwen)

Qwen3 TTS Instruct Flash

Speech controlled through natural-language direction
Fine control over pace, emotion, and style

audio outstreaming

Task rate

$0.1437

Billing

Per task

see pricing

Video generationImage generation

Kuaishou (Kling)

Kling 3.0

Kling 3 tier for longer multi-shot storytelling
Stronger narrative continuity and camera planning

video outputimage output

Task rate

$0.1904

per second

Billing

Per task

see pricing

Google DeepMind (Gemini)

Gemini Embedding 2

Knowledge search

Unified embeddings for text, images, video, and audio
Cross-modal retrieval in one semantic space

8K ctxmultimodal

Lowest rate

$0.18

per 1M tokens

Billing

Metered

see pricing

Google DeepMind (Gemini)

Gemini 2.5 Flash TTS Preview

Responsive Gemini speech tier for frequent generation
Controllable single- and two-speaker delivery

8K ctxaudio out

Lowest rate

$0.45 / $9.00

per 1M tokens

Billing

Metered

see pricing

Google DeepMind (Gemini)

Gemini 2.5 Pro TTS Preview

Quality-first Gemini speech for polished output
Suited to long-form and expressive narration

8K ctxaudio out

Lowest rate

$0.90 / $18.00

per 1M tokens

Billing

Metered

see pricing

ByteDance (Doubao)

Seedance 1.5 Pro

Native joint video-and-audio generation
Dialogue, lip sync, and continuous narrative

video outputaudio out

Lowest rate

$2.70 / $2.70

per 1M tokens

Billing

Metered

see pricing

Alibaba (Qwen)

FunASR

Speech to text

General-purpose speech recognition
Built for file transcription at scale

audio in

Task rate

$0.000028

Billing

Per task

see pricing

Alibaba (Qwen)

Qwen3 ASR Flash

Speech to text

Fast recognition for short audio
Multilingual transcription with emotion cues

audio in

Task rate

$0.000028

Billing

Per task

see pricing

Alibaba (Qwen)

Qwen3 ASR Flash FileTrans

Speech to text

Long-file transcription for recordings up to 12 hours
Word-level timestamps for searchable audio

audio in

Task rate

$0.000028

Billing

Per task

see pricing

OpenAI

GPT Image 2

Precise instruction following with high-fidelity references
Strong text rendering for professional editing

Task rate

$0.0054

per image

Billing

Per task

see pricing

Kuaishou (Kling)

Kling 2.1

Kling model for expressive motion and camera work
Built for high-control video production

Task rate

$0.012

per image

Billing

Per task

see pricing

Alibaba (Qwen)

Qwen Image 2.0

Balanced Qwen image generation and editing
Combines text layout quality with everyday production speed

Task rate

$0.0229

per Piece

Billing

Per task

see pricing

Unitper Piece

Alibaba (Qwen)

WAN 2.7 Image

Wan 2.7 image generation
Controllable visual generation

Task rate

$0.0229

per Piece

Billing

Per task

see pricing

Unitper Piece

Google DeepMind (Gemini)

Veo 3.1 Lite

Video generation
Built for short-form content

video output

Task rate

$0.027

per second

Billing

Per task

see pricing

MiniMax

MiniMax Music 2.0

Music generation

Fast full-song generation from style and lyrics
Cost-oriented tier for drafts and batches

audio out

Task rate

$0.0285

per clip

Billing

Per task

see pricing

Unitper clip

Alibaba (Qwen)

Qwen Image 3.0

Fast production image generation
Generation and editing

Task rate

$0.03

per image

Billing

Per task

see pricing

Vidu

Vidu q2-turbo

Speed- and cost-oriented Vidu Q2 tier
Fast image animation and shot extension

video output

Task rate

$0.0331

per second

Billing

Per task

see pricing

ByteDance (Doubao)

Seedream 4.0

Unified Seedream generation and editing
Multi-reference consistency for general visual work

Task rate

$0.0338

per image

Billing

Per task

see pricing