LLM API Cost Calculator
Estimate your monthly API bill across Claude Opus 4.8, GPT-5.6 (Sol / Terra / Luna) and Grok 4.5 at official list prices — including the effect of prompt caching — plus per-unit costs for image and video models. All text-model rates below are the vendors' published list prices (July 2026).
Via byesu you pay a fraction of this list-price bill — billing uses transparent multipliers (input 1x · output 5x · cache hit 0.1x) and the console shows your live billed price next to the official list price.
See live prices in the console →List prices are the vendors’ published API prices (July 2026) and may change; media prices are byesu’s live per-unit prices.
How to read the numbers
- Input / output split. Output tokens cost 3–6× more than input on every model here. Coding agents are output-heavy, so verbosity matters as much as the rate card: according to xAI, Grok 4.5 emits ~14k output tokens on task batches where Claude Opus 4.8 emits ~67k.
- Cache hit rate. Cached input tokens bill at roughly 10% of the input rate (25% for Grok 4.5). Long system prompts, big repo contexts and multi-turn agent loops routinely reach 50–80% cache hit rates, which is why the "with caching" number drops so hard.
- Tier choice on GPT-5.6. Terra delivers most of Sol's benchmark performance at half the price; Luna is the volume tier. See the full Sol vs Terra vs Luna guide.
What you actually pay on byesu
byesu is an AI API gateway: one key covers all the models above through an OpenAI-compatible endpoint plus an Anthropic-native /v1/messages endpoint. Billing is pay-as-you-go with transparent multipliers:
| Billing item | Multiplier |
|---|---|
| Input | 1x |
| Output | 5x |
| Cache hit | 0.1x |
You pay a fraction of the official list-price bill shown by the calculator — the console shows your live billed price next to the official list price, so you can verify it per model before committing volume. The cache-hit structure matches the official input : output : cache-hit = 1 : 5 : 0.1 pattern, so the caching savings estimated above carry over.
Top up with USDT, Alipay or WeChat Pay. Media tasks (image / video) are auto-refunded on failure.
Related pages
- Grok 4.5 API — access & pricing
- GPT-5.6 Sol vs Terra vs Luna
- Claude Opus 4.8 API
- Full pricing & billing overview
FAQ
How is LLM API usage billed?
Per token: one rate per million input tokens, a higher rate per million output tokens. Official list prices used by this calculator: Claude Opus 4.8 $5/$25, GPT-5.6 Sol $5/$30, Terra $2.5/$15, Luna $1/$6, Grok 4.5 $2/$6 per million tokens.
How much does prompt caching save?
Cached input tokens bill at about 10% of the input rate (25% on Grok 4.5). At a 60% cache hit rate, the input side of your bill drops by more than half — move the slider above to see the effect on your own volumes.
Does the calculator show what I would pay on byesu?
For text models it shows the official list-price bill; on byesu you pay a fraction of that, and the console shows the live billed price next to the official list price. Media prices shown are byesu's live per-unit prices ($0.01/image for gpt-image-2, $0.01875/sec for Veo 3.1, etc.).