Pricing

Simple, usage-based pricing. Pay only for what you use.

How It Works

Purchase Credits

Add funds to your account via Stripe. Your balance is available immediately.

Use the API

Make requests to any supported model. Costs are calculated per token.

Track Usage

Monitor your spending in real-time. Get detailed analytics on your usage.

Text Model Pricing

Pricing per 1,000,000 tokens for each model

ModelDescriptionInput CostOutput Cost
mistral-voxtral-mini-3b-2507Japan: Mistral's voxtral-mini-3b-2507¥8.4 / 1M tokens¥8.4 / 1M tokens
anthropic-claude-opus-4-8Global: Claude Opus 4.8 - Most intelligent, best for agents and coding¥893.5 / 1M tokens¥4,467.5 / 1M tokens
anthropic-claude-opus-4-7Global: Claude Opus 4.7 - Most intelligent, best for agents and coding¥840 / 1M tokens¥4,200 / 1M tokens
anthropic-claude-opus-4-6Global: Claude Opus 4.6 - Most intelligent, best for agents and coding¥850 / 1M tokens¥4,200 / 1M tokens
anthropic-claude-sonnet-4-6Japan: Claude Sonnet 4.6 - Best balance of speed and intelligence¥554.4 / 1M tokens¥2,772 / 1M tokens
anthropic-claude-haiku-4-5Japan: Claude Haiku 4.5 - Fastest with near-frontier intelligence¥184.8 / 1M tokens¥924 / 1M tokens
minimax-m2.7Global: Minimax's M2.7¥53.61 / 1M tokens¥214.44 / 1M tokens
glm-5Global: Z.ai's GLM-5¥178.7 / 1M tokens¥571.84 / 1M tokens
llm-jp-3.1-8x13b-instruct4Japan: Japanese language model optimized for instruction following¥16 / 1M tokens¥79 / 1M tokens
japan-gemma-4-31bJapan: Google's Gemma 4 31B - hosted and run in Japan¥26 / 1M tokens¥101 / 1M tokens
japan-qwen3.6-35bJapan: Qwen 3.6 35B-A3B - hosted and run in Japan¥32 / 1M tokens¥158 / 1M tokens
japan-kimi-k2.6Japan: Moonshot AI's Kimi K2.6 - hosted and run in Japan¥64 / 1M tokens¥316 / 1M tokens
japan-kimi-k2.7-codeJapan: Moonshot AI's Kimi K2.7 Code - hosted and run in Japan¥55 / 1M tokens¥530 / 1M tokens
gpt-oss-120bJapan: gpt-oss-120b¥16 / 1M tokens¥79 / 1M tokens
minimax-m3Global: MiniMax's M3¥53.61 / 1M tokens¥214.44 / 1M tokens
qwen3.6-27bGlobal: Qwen 3.6 27B¥80.415 / 1M tokens¥482.49 / 1M tokens
qwen3.8-27bGlobal: Qwen 3.8 27B¥98.285 / 1M tokens¥589.71 / 1M tokens
kimi-k2.6Global: Moonshot AI's Kimi K2.6¥169.765 / 1M tokens¥714.8 / 1M tokens
glm-4.7Global: Z.ai's GLM 4.7¥107.22 / 1M tokens¥393.14 / 1M tokens
glm-4.7-flashGlobal: Z.ai's GLM 4.7 Flash¥10.722 / 1M tokens¥71.48 / 1M tokens
glm-5.1Global: Z.ai's GLM 5.1¥250.18 / 1M tokens¥786.28 / 1M tokens
mimo-v2.5Global: Xiaomi's MiMo-V2.5¥25.018 / 1M tokens¥50.036 / 1M tokens
mimo-v2.5-proGlobal: Xiaomi's MiMo-V2.5-Pro¥77.7345 / 1M tokens¥155.469 / 1M tokens
deepseek-v4-flashGlobal: DeepSeek's V4 Flash¥16.083 / 1M tokens¥32.166 / 1M tokens
deepseek-v4-flash-0731Global: DeepSeek's V4 Flash 0731¥25.018 / 1M tokens¥50.036 / 1M tokens
deepseek-v4-proGlobal: DeepSeek's V4 Pro¥341.317 / 1M tokens¥684.421 / 1M tokens
glm-5.2Global: Z.ai's GLM 5.2¥250.18 / 1M tokens¥786.28 / 1M tokens
glm-5.3Global: Z.ai's GLM 5.3¥250.18 / 1M tokens¥786.28 / 1M tokens
kimi-k3Global: Moonshot AI's Kimi K3¥536.1 / 1M tokens¥2,680.5 / 1M tokens
qwen3.7-maxGlobal: Qwen 3.7 Max¥263.5825 / 1M tokens¥790.7475 / 1M tokens
qwen3.8-maxGlobal: Qwen 3.8 Max¥357.4 / 1M tokens¥1,072.2 / 1M tokens
grok-4.5Global: xAI's Grok 4.5¥357.4 / 1M tokens¥1,072.2 / 1M tokens
grok-4.6Global: xAI's Grok 4.6¥357.4 / 1M tokens¥1,072.2 / 1M tokens
gpt-5.6-solGlobal: OpenAI's GPT-5.6 Sol¥893.5 / 1M tokens¥5,361 / 1M tokens
gpt-5.6-terraGlobal: OpenAI's GPT-5.6 Terra¥357.4 / 1M tokens¥2,144.4 / 1M tokens
gpt-5.6-lunaGlobal: OpenAI's GPT-5.6 Luna¥35.74 / 1M tokens¥214.44 / 1M tokens
gemini-3.5-flashGlobal: Google's Gemini 3.5 Flash¥268.05 / 1M tokens¥1,608.3 / 1M tokens
gemini-3.7-flashGlobal: Google's Gemini 3.7 Flash¥134.025 / 1M tokens¥670.125 / 1M tokens
inklingGlobal: Thinking Machines' Inkling¥178.7 / 1M tokens¥723.735 / 1M tokens
anthropic-claude-opus-5Global: Claude Opus 5 - Most intelligent, best for agents and coding¥893.5 / 1M tokens¥4,467.5 / 1M tokens
anthropic-claude-sonnet-5Global: Claude Sonnet 5 - Best balance of speed and intelligence¥357.4 / 1M tokens¥1,787 / 1M tokens
anthropic-claude-fable-5Global: Claude Fable 5 - Anthropic's flagship Mythos-class model¥1,787 / 1M tokens¥8,935 / 1M tokens
muse-glimmer-30bGlobal: Meta's Muse Glimmer 30B¥62.545 / 1M tokens¥268.05 / 1M tokens
gemini-flash-lite-freeGlobal: Google's Gemini Flash Lite - free to try¥0 / 1M tokens¥0 / 1M tokens
random-freeGlobal: Random - free to try¥0 / 1M tokens¥0 / 1M tokens

Image Model Pricing

Per-image pricing for image generation models

ModelDescriptionCost
google-nano-banana-2Global: Google Nano Banana 2 - fast, high-quality image generation¥12.23 / image
bytedance-seedream-4.5Global: Generate images¥7.15 / image
z-image-turboGlobal: Generate realistic images¥1.68 / image
lustify-sdxlGlobal: Generate animated images¥1.68 / image

Video Model Pricing

Flat per-video pricing for video generation models

ModelSpecsCost
Seedance 2.0 Fast8s · 1280x720 · with audio¥250 / video

Using FastMetal at your company

Billing in yen, qualified invoices, and a prepaid balance that caps spend — with the questions procurement usually asks answered in one place.

See the business page

Ready to Start?

Create an account and purchase your first credits.

Get Started