Pricing
Pay only for what you use across all 86 models — one balance, one bill. No subscription, no seat fees, no minimums. Language models bill per token; image, video, and audio bill per unit produced. Start with $0.50 in free credit, no card required.
$0.50 free credit
On signup, no credit card. Enough to test every model class.
Cached input at 10%
Repeated prompt prefixes are billed at 10% of the input rate — up to 90% off.
Refund on failure
Image, video, and audio jobs that fail upstream are refunded in full.
Language & reasoning models
Billed per 1M tokens. Cached input is 10% of the input rate where supported.
| Model | Creator | Context | Input /1M | Cached /1M | Output /1M |
|---|---|---|---|---|---|
| EmbeddingGemma 300M | 2K | $0.0027 | $0.00027 | $0.00 | |
| Qwen3 Embedding 8B | Alibaba | 33K | $0.0135 | $0.00135 | $0.00 |
| Qwen 3.7 Flash | Alibaba | 1M | $0.0405 | $0.00405 | $0.1755 |
| GLM 4.7 Flash | Zhipu AI | 203K | $0.081 | $0.0081 | $0.54 |
| GLM 4.5 Air | Zhipu AI | 131K | $0.176 | $0.0176 | $1.148 |
| Gemma 4 31B | 128K | $0.189 | $0.0189 | $0.54 | |
| DeepSeek V4 Flash | DeepSeek | 1M | $0.189 | $0.0189 | $0.378 |
| GPT-OSS 120B | OpenAI | 128K | $0.203 | $0.0203 | $0.81 |
| Step 3.7 Flash | StepFun | 256K | $0.27 | $0.027 | $1.553 |
| Qwen 3 32B | Alibaba | 33K | $0.392 | $0.0392 | $0.81 |
| Gemini 3.5 Flash Lite | 1M | $0.405 | $0.0405 | $3.375 | |
| MiniMax M2.5 | MiniMax | 197K | $0.405 | $0.0405 | $1.62 |
| MiniMax M3 | MiniMax | 1M | $0.405 | $0.0405 | $1.62 |
| Gemini 2.5 Flash | 1M | $0.405 | $0.0405 | $3.375 | |
| MiniMax M2.7 | MiniMax | 205K | $0.405 | — | $1.62 |
| Qwen 3.7 Plus | Alibaba | 1M | $0.432 | $0.0432 | $1.728 |
| Qwen 3.6 Plus | Alibaba | 131K | $0.675 | $0.0675 | $4.05 |
| Qwen 3 Coder | Alibaba | 131K | $0.675 | $0.0675 | $2.16 |
| Gemini 3 Flash | 1M | $0.675 | $0.0675 | $4.05 | |
| Nemotron 3 Ultra 550B | NVIDIA | 1M | $0.675 | $0.0675 | $3.375 |
| Kimi K2.5 | Moonshot | 262K | $0.675 | — | $3.78 |
| DeepSeek R1 | DeepSeek | 64K | $0.743 | — | $2.957 |
| DeepSeek V3 | DeepSeek | 160K | $0.81 | — | $2.295 |
| Llama 3.3 70B | Meta | 128K | $1.188 | $0.1188 | $1.188 |
| Kimi K2.7 Code | Moonshot | 262K | $1.283 | — | $5.40 |
| Kimi K2.6 | Moonshot | 262K | $1.283 | — | $5.40 |
| Claude Haiku 4.5 | Anthropic | 200K | $1.35 | $0.135 | $6.75 |
| Grok Build | xAI | 256K | $1.35 | $0.135 | $2.70 |
| Sonar | Perplexity | 127K | $1.35 | $0.135 | $1.35 |
| Grok 4.3 | xAI | 1M | $1.688 | $0.1688 | $3.375 |
| GLM 5.2 | Zhipu AI | 1M | $1.89 | $0.189 | $5.94 |
| GLM 5.1 | Zhipu AI | 203K | $1.89 | $0.189 | $5.94 |
| Qwen 3.7 Max | Alibaba | 1M | $1.991 | $0.1991 | $5.974 |
| Gemini 3.6 Flash | 1M | $2.025 | $0.2025 | $10.125 | |
| Gemini 3.5 Flash | 1M | $2.025 | $0.2025 | $12.15 | |
| DeepSeek V4 Pro | DeepSeek | 1M | $2.349 | $0.2349 | $4.698 |
| GPT-5.6 Terra | OpenAI | 1M | $2.70 | $0.27 | $16.20 |
| Kimi K3 | Moonshot | 1M | $4.05 | $0.405 | $20.25 |
| Claude Sonnet 4.6 | Anthropic | 1M | $4.05 | $0.405 | $20.25 |
| Sonar Pro | Perplexity | 200K | $4.05 | $0.405 | $20.25 |
| Claude Opus 4.7 | Anthropic | 1M | $6.75 | $0.675 | $33.75 |
Image generation
Billed per generated image. You only pay for successful renders.
| Model | Creator | Price |
|---|---|---|
| MiniMax Image 01 | MiniMax | $0.005 / image |
| Imagen 4 Fast | $0.027 / image | |
| FLUX.2 Pro | Black Forest Labs | $0.0405 / image |
| Nano Banana | $0.046 / image | |
| Nano Banana 3 Flash (preview) | $0.046 / image | |
| FLUX.1 Kontext Pro | Black Forest Labs | $0.054 / image |
| Recraft V4 | Recraft | $0.054 / image |
| Recraft V3 | Recraft | $0.054 / image |
| Imagen 4 | $0.054 / image | |
| FLUX 1.1 Pro Ultra | Black Forest Labs | $0.081 / image |
| Imagen 4 Ultra | $0.081 / image | |
| GPT Image 2 | OpenAI | $0.081 / image |
| Ideogram V3 | Ideogram | $0.108 / image |
| Recraft V4 Vector | Recraft | $0.108 / image |
| Recraft V4 Pro | Recraft | $0.3375 / image |
| Recraft V4 Vector Pro | Recraft | $0.405 / image |
Video generation
Billed per second of generated video, or per clip. Failed jobs are refunded.
| Model | Creator | Price |
|---|---|---|
| Kling 2.5 Pro | Kuaishou | $0.0945 / sec of video |
| Veo 3 Fast | $0.135 / sec of video | |
| ElevenLabs Music | ElevenLabs | $0.135 / sec of video |
| Hailuo 02 (512p) | MiniMax | $0.14 / clip |
| Kling 3 Pro | Kuaishou | $0.1512 / sec of video |
| Kling 3 Pro (Audio) | Kuaishou | $0.2268 / sec of video |
| Seedance 2 Fast | ByteDance | $0.3266 / sec of video |
| Seedance 2 Pro | ByteDance | $0.4096 / sec of video |
| Hailuo 02 (768p) | MiniMax | $0.42 / clip |
| Veo 3 | $0.54 / sec of video | |
| Hailuo 02 (1080p) | MiniMax | $0.78 / clip |
Audio — speech, transcription, music & sound
Text-to-speech bills per 1K characters, transcription and understanding per minute, music per track, sound effects per generation.
| Model | Creator | Type | Price |
|---|---|---|---|
| Whisper Large v3 Turbo | OpenAI | audio | $0.0009 / min |
| GPT-4o mini Transcribe | OpenAI | audio | $0.004 / min |
| Gemini 3 Flash (Audio) | audio | $0.0026 / min | |
| GPT Realtime Translate | OpenAI | audio | $0.0459 / min |
| Gemini 2.5 Flash Native Audio | audio | $0.0389 / min | |
| Gemini 3.1 Flash Live | audio | $0.0389 / min | |
| Gemini 3.5 Live Translate | audio | $0.0635 / min | |
| ElevenLabs Multilingual v2 | ElevenLabs | audio | $0.405 / 1K chars |
| ElevenLabs v3 | ElevenLabs | audio | $0.405 / 1K chars |
| ElevenLabs Flash v2.5 | ElevenLabs | audio | $0.2025 / 1K chars |
| ElevenLabs Turbo v2.5 | ElevenLabs | audio | $0.2025 / 1K chars |
| ElevenLabs Sound Effects | ElevenLabs | audio | $0.027 / generation |
| MiniMax Speech HD | MiniMax | audio | $0.14 / 1K chars |
| MiniMax Speech Turbo | MiniMax | audio | $0.09 / 1K chars |
| MiniMax Music | MiniMax | audio | $0.045 / track |
| MiniMax Music Pro | MiniMax | audio | $0.21 / track |
| MiniMax Voice Clone | MiniMax | audio | $2.10 / generation |
| MiniMax Voice Design | MiniMax | audio | $4.20 / generation |
How billing works
- • Every request places a hold at the quoted price and settles on the actual usage returned. If the request fails, the hold is released — you are never charged for an error.
- • Prompt caching is automatic for supported language models: repeated system prompts and prefixes are billed at 10% of the input rate.
- • Rate limits scale by tier. See rate limits and the full pricing reference in the docs.
- • Browse every model with live performance data on the models page, or compare options on the comparison page.
$0.50 free credit on signup. No credit card required.
Get your free API key →