LLM API Cost Calculator (October 2026): Claude vs GPT-6 vs Gemini Official Prices
An LLM API cost calculator comes down to one formula and a current price list. This guide gives you both: ten models from Anthropic, OpenAI and Google, with list prices copied from each official pricing page on 2026-10-18. It compares cost only, not quality.
Three details can change your bill: higher rates for long prompts on some models, and two Gemini introductory prices that end on fixed dates.
Quick answer: the formula and the price table
All three providers list prices in US dollars per 1 million tokens. So the cost of one request is:
(input tokens / 1,000,000 x input price) + (output tokens / 1,000,000 x output price)
Multiply the result by your requests per day and your days per month. The calculator below uses 30 days by default.
| Model | Input per 1M tokens | Output per 1M tokens | Note |
|---|---|---|---|
| Claude Fable 5.1 | $10 | $50 | Anthropic |
| Claude Opus 5.5 | $4 | $20 | Anthropic, released September 22, 2026 |
| Claude Sonnet 5.5 | $2 | $10 | Anthropic, released September 28, 2026 |
| Claude Haiku 5.5 | $0.10 | $0.50 | Prompts up to 100,000 tokens; released October 7, 2026 |
| GPT-6 Astra | $10.00 | $50.00 | OpenAI |
| GPT-6.1 Sol | $2.00 | $10.00 | OpenAI |
| GPT-6 Luna | $0.10 | $0.50 | OpenAI |
| Gemini 4 Argon | $2.00 | $10.00 | Introductory price through November 30, 2026 |
| Gemini 3.8 Flash | $0.75 | $3.75 | Introductory price through December 31, 2026 |
| Gemini 3.5 Flash-Lite | $0.30 | $2.50 |
These are Standard list prices. On Google's pricing page, Gemini output prices include thinking tokens.
Calculator and worked example
Enter your average input tokens and output tokens per request, your requests per day, and your days per month.
LLM API cost calculator
1. Model
2. Input tokens per request (prompt)
= 2,000 tokens
3. Output tokens per request (reply)
= 500 tokens
4. Requests per day
= 1,000 requests
5. Days per month
Standard list prices per 1 million tokens from the official Anthropic, OpenAI and Google pricing pages, checked 2026-10-18. Batch discounts, prompt caching, tools, taxes and free tiers are not included. Reasoning or thinking tokens count as output if the provider bills them that way. OpenAI's 272K long-context threshold is treated as 272,000 input tokens.
To see the formula by hand, take 2,000 input tokens and 500 output tokens on Claude Sonnet 5.5. Input costs $0.004 and output costs $0.005, so one request is $0.009. At 1,000 requests a day for 30 days, that is $9 a day and $270 a month.
Here is the same workload (2,000 in, 500 out, 1,000 requests a day, 30 days) on every model in the table:
| Model | Monthly cost |
|---|---|
| Claude Haiku 5.5 | $13.50 |
| GPT-6 Luna | $13.50 |
| Gemini 3.5 Flash-Lite | $55.50 |
| Gemini 3.8 Flash (introductory price) | $101.25 |
| Claude Sonnet 5.5 | $270 |
| GPT-6.1 Sol | $270 |
| Gemini 4 Argon (introductory price) | $270 |
| Claude Opus 5.5 | $540 |
| Claude Fable 5.1 | $1,350 |
| GPT-6 Astra | $1,350 |
Your own workload will differ, so replace these token counts and request volumes with your own.
Where the price changes: long prompts and Gemini deadlines
Long prompts on Claude Haiku 5.5
Claude Haiku 5.5 costs $0.10 input and $0.50 output per million tokens for prompts up to 100,000 tokens. If a prompt is over 100,000 tokens, the whole request is billed at $0.50 input and $2.50 output per million tokens. Cache reads and writes count toward the prompt length.
Anthropic says Claude 4.6 and later models, except Claude Haiku 5.5, include the full 1M-token context window at standard pricing. For Fable 5.1, Opus 5.5 and Sonnet 5.5, a long prompt pays the same per-token rate as a short one. Each of Anthropic's four current models has a 1M-token context window and up to 128K output tokens.
Long prompts on OpenAI GPT-6 models
For GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna, prompts with more than 272K input tokens are priced at 2x the input rate and 1.5x the output rate for the full request. That applies to the whole request, not only the tokens past the threshold.
With this rule, GPT-6.1 Sol costs $4.00 input and $15.00 output per million tokens. GPT-6 Astra costs $20.00 input and $75.00 output per million tokens. The calculator treats the threshold as 272,000 input tokens. All three models have a context window of 1,050,000 tokens and up to 128,000 output tokens.
Gemini introductory prices that end
Two Gemini prices in the table are temporary, and Google lists the dates:
- Gemini 4 Argon: $2.00 input and $10.00 output per million tokens through November 30, 2026. From December 1, 2026, it costs $4.00 input and $20.00 output.
- Gemini 3.8 Flash: $0.75 input and $3.75 output through December 31, 2026. From January 1, 2027, it costs $1.50 input and $7.50 output.
If you are budgeting past those dates, use the later prices rather than the ones in the table. Gemini 3.8 Flash is a stable model with an input limit of 1,048,576 tokens.
What the calculator leaves out
The calculator uses Standard list prices checked on 2026-10-18 and a default of 30 days per month. Batch discounts, prompt caching, tools and taxes are not included.
One discount is confirmed in the sources used here: Anthropic's Batch API gives a 50% discount on both input and output tokens. Whether it suits your workload depends on your situation, so check Anthropic's documentation. For any other discount or billing detail from OpenAI or Google, check their official pricing pages.
Prices can change. Before you commit to a budget, compare these figures with each provider's current pricing page.
FAQ
How do I calculate LLM API cost?
Divide input tokens by 1 million and multiply by the input price. Do the same for output tokens with the output price, then add the two. For example, 2,000 input and 500 output tokens on Claude Sonnet 5.5 cost $0.009.
Which is cheapest: Claude Haiku 5.5, GPT-6 Luna or Gemini 3.5 Flash-Lite?
On list price, Claude Haiku 5.5 and GPT-6 Luna both charge $0.10 input and $0.50 output per million tokens. Gemini 3.5 Flash-Lite charges $0.30 input and $2.50 output. Haiku 5.5 rises to $0.50 and $2.50 when a prompt is over 100,000 tokens.
Do long prompts cost more?
For some models, yes. OpenAI's GPT-6 models charge 2x input and 1.5x output for the full request when the prompt has more than 272K input tokens. Claude Haiku 5.5 charges a higher rate above 100,000 prompt tokens. Anthropic says its other current models keep standard pricing across the 1M-token context window.
Will Gemini API prices go up?
Google lists dated increases. Gemini 4 Argon goes from $2.00 and $10.00 to $4.00 and $20.00 per million tokens on December 1, 2026. Gemini 3.8 Flash goes from $0.75 and $3.75 to $1.50 and $7.50 on January 1, 2027.
Related guides
Sources
- Anthropic (Claude Platform Docs) — Pricing
- Anthropic (Claude Platform Docs) — Models overview
- Anthropic (Claude Platform Docs) — Claude Opus 5.5
- Anthropic (Claude Platform Docs) — Claude Haiku 5.5
- Anthropic (Claude Platform Docs) — Claude Fable 5.1
- Anthropic — Introducing Claude Sonnet 5.5
- OpenAI API Docs — Pricing
- OpenAI API Docs — GPT-6.1 Sol Model
- OpenAI API Docs — GPT-6 Astra Model
- OpenAI API Docs — GPT-6 Luna Model
- OpenAI API Docs — Compare models
- Google AI for Developers — Gemini Developer API pricing
- Google AI for Developers — What's new in Gemini 4 Argon
- Google AI for Developers — What's new in Gemini 3.8 Flash
- Google AI for Developers — What's new in Gemini 3.6 Flash and 3.5 Flash-Lite
- Google (The Keyword) — Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber
- ComputingForGeeks — Gemini 4 Argon released: features, benchmarks and pricing
Facts last verified: 2026-10-18
Comments
Post a Comment