AIin1Portal

Pricing

Billing Model

AIone uses pay-as-you-go billing based on actual token consumption.

  • Input Tokens: Content sent to the model (system prompt + user message + conversation history)
  • Output Tokens: Content generated by the model
  • Billing Unit: Per million tokens by default
  • Currency: All prices are displayed in USD in the console

Commonly Available Models

Models evolve rapidly, so this document does not label any specific model as "the latest." The table below lists commonly available model families as a reference. For current availability, pricing, and any applicable discounts, always refer to the console.

Family Commonly Available Models
Claude claude-haiku-5-5, claude-opus-4-1-20250805, claude-opus-4-20250514, claude-sonnet-4-20250514, claude-3-7-sonnet-20250219, claude-3-5-haiku-20241022
GPT gpt-5.4, gpt-5.2, gpt-5, gpt-5-mini, gpt-5-nano
Gemini gemini-nano-banana-2.1, gemini-3.1-pro, gemini-3.1-flash-lite-preview, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite

For real-time pricing, visit the My Pricing page.

The model list in this document is not a fixed price sheet. Models and pricing may change based on model-vendor capabilities, platform strategy, and enterprise discounts.

Haiku 5.5: pricing above 100K input tokens

For claude-haiku-5-5, the pricing tier depends on total input for the entire request:

Total input tokens Token rates for this request
≤ 100,000 Base rates shown in the console
> 100,000 All input, output, cache write, and cache read rates are 5 times the base rates

Total input = uncached input + cache writes + cache reads. One-hour cache writes are a subset of cache writes; do not add them twice. Exactly 100,000 tokens stays in the base tier. Above the threshold, the higher rates apply to the whole request, not just the excess tokens.

For example, 80,000 uncached input + 30,000 cache reads = 110,000, so the higher tier applies even though new input alone is below 100K. Organization discounts and selected service-channel multipliers still apply as shown in the console. See the Haiku 5.5 guide for parameters and examples.

Nano Banana 2.1: image output billing

For gemini-nano-banana-2.1, image output tokens per image are 1,120 at 1K, 1,680 at 2K, and 3,780 at 4K. Image output uses the image rate; text and thinking output use the text rate. Input is billed separately. Do not use an older Nano model's 4K tier to calculate the new model's cost. See Gemini image generation for request parameters.

Enterprise Discounts

Enterprise customers receive a unified discount rate. Discounted prices are visible on the My Pricing page in the console.

For custom quotes, payment terms, or higher quotas, please contact our sales team through a support ticket.

Cost Optimization Tips

  • For high-frequency, standardized tasks, prefer gpt-5-nano, gpt-5-mini, gemini-2.5-flash-lite, or gemini-2.5-flash
  • For code generation and complex analysis where quality matters, consider claude-sonnet-4-20250514, claude-3-7-sonnet-20250219, or gpt-5.2
  • For deep reasoning or mission-critical workloads, consider claude-opus-4-1-20250805, claude-opus-4-20250514, gpt-5.4, or gemini-3.1-pro
  • Combine Prompt Caching with appropriate max_tokens settings to further reduce per-task costs

Viewing Usage