AI pricing calculator
Compare current standard API rates. The estimate uses your request volume, input size, and output size. It does not include cache, batch, or fast-mode discounts.
Model APIs / checked 17 October 2026
Select Use Case
Configuration
1,000 tokens
500 tokens
1,000 / day (30,000 / month)
Estimated Monthly Cost
Based on 30,000 requests/month with 1,000 input + 500 output tokens each
Claude Haiku 5.5
Lowest estimateAnthropic • High-volume tasks. Rate is for prompts up to 100,000 tokens
$0.10 in / $0.50 out per 1M tokens
Input: $3.00 | Output: $7.50 | Per request: <$0.01
$10.50
/month
GPT-6 Luna
OpenAI • Lowest listed rate in the current flagship family
$0.10 in / $0.50 out per 1M tokens
Input: $3.00 | Output: $7.50 | Per request: <$0.01
$10.50
/month
Gemini 3.5 Flash-Lite
Google • Cost-efficient text, image, video, and audio rate
$0.30 in / $2.50 out per 1M tokens
Input: $9.00 | Output: $37.50 | Per request: <$0.01
$46.50
/month
Grok Build 0.1
Grok • Coding model. Rate is for prompts under 200,000 tokens
$1.00 in / $2.00 out per 1M tokens
Input: $30.00 | Output: $30.00 | Per request: <$0.01
$60.00
/month
Grok 4.3
Grok • Lower text tier. Rate is for prompts under 200,000 tokens
$1.25 in / $2.50 out per 1M tokens
Input: $37.50 | Output: $37.50 | Per request: <$0.01
$75.00
/month
Gemini 3.8 Flash
Google • Paid standard rate through 31 December 2026
$0.75 in / $3.75 out per 1M tokens
Input: $22.50 | Output: $56.25 | Per request: <$0.01
$78.75
/month
Grok 4.7
Grok • Latest flagship. Rate is for prompts under 200,000 tokens
$2.00 in / $6.00 out per 1M tokens
Input: $60.00 | Output: $90.00 | Per request: <$0.01
$150.00
/month
Claude Sonnet 5.5
Anthropic • Speed and intelligence for coding and agents
$2.00 in / $10.00 out per 1M tokens
Input: $60.00 | Output: $150.00 | Per request: <$0.01
$210.00
/month
GPT-6.1 Sol
OpenAI • Standard short-context rate
$2.00 in / $10.00 out per 1M tokens
Input: $60.00 | Output: $150.00 | Per request: <$0.01
$210.00
/month
Gemini 3.1 Pro Preview
Google • Preview. Standard rate for prompts up to 200,000 tokens
$2.00 in / $12.00 out per 1M tokens
Input: $60.00 | Output: $180.00 | Per request: <$0.01
$240.00
/month
Claude Opus 5.5
Anthropic • Long-running agentic coding and knowledge work
$4.00 in / $20.00 out per 1M tokens
Input: $120.00 | Output: $300.00 | Per request: $0.0140
$420.00
/month
Claude Fable 5.1
Anthropic • Demanding reasoning and long-horizon agent work
$10.00 in / $50.00 out per 1M tokens
Input: $300.00 | Output: $750.00 | Per request: $0.0350
$1050.00
/month
GPT-6 Astra
OpenAI • Latest model. Standard short-context rate
$10.00 in / $50.00 out per 1M tokens
Input: $300.00 | Output: $750.00 | Per request: $0.0350
$1050.00
/month
* Standard API rates checked 17 October 2026. OpenAI figures are the short-context rate, for inputs at or under 272,000 tokens. Haiku 5.5 uses the rate for prompts at or under 100,000 tokens. Gemini 3.8 Flash uses the paid standard rate through 31 December 2026. Grok figures are for prompts under 200,000 tokens. Grok 4.6 and Grok 4.5 use the same under-200,000-token rate as Grok 4.7, so they are not listed separately. Cached input, batch, priority, long-context surcharges, and separate thinking-token charges are not included.
Understanding AI API pricing
AI model pricing is typically based on tokens, which are pieces of text—roughly four characters or 0.75 words in English. Pricing is usually quoted per million tokens, with separate rates for input and output.
Key pricing factors
- Input tokens—system prompts, user messages, and context sent to the model
- Output tokens—the text the model generates in response
- Model tier—more capable models generally cost more
- Context window—larger windows admit more input and can increase costs
Cost optimization tips
- Choose the right model—use smaller models for routine tasks and reserve more capable models for complex reasoning
- Optimize prompts—concise, well-structured prompts reduce input token costs
- Limit output length—set output limits when the task has a predictable response shape
- Use caching—reuse stable prompt prefixes and common responses where providers support it
- Batch requests—some providers offer discounted asynchronous processing
Provider comparison
| Provider | Strengths | Best for |
|---|---|---|
| Anthropic | Long context, coding, safety | Complex analysis and code generation |
| OpenAI | Ecosystem and multimodal models | General-purpose and vision tasks |
| Large context and multimodal inputs | Document processing and long content | |
| Grok | Current text and coding models | Agent work and code, at the published token rate |
Rates were read from the provider pricing pages on 17 October 2026: OpenAI, Anthropic, Google, and Grok. Check those pages before you treat an estimate as a bill.