GPT-4o API Pricing 2026: $2.5/M Tokens
Complete per-million-token pricing for GPT-4o from OpenAI, including cached and batch rates, cost-per-1K/1M/10M breakdowns, a monthly cost calculator, and a head-to-head comparison with Claude Sonnet 4, Gemini 2.5 Pro, DeepSeek-V3. For full cross-provider rankings, open the main calculator.
GPT-4o headline pricing
Input
$2.50 /M tokens
Output
$10.00 /M tokens
Cached input
$1.25 /M (50% off)
Batch (50% off)
$1.25 in / $5.00 out /M
Context · capabilities
128K context · Chat, Code, Vision, Reasoning · benchmark 9/10
GPT-4o cost per 1K, 1M and 10M tokens
Standard on-demand USD rates. Input assumes prompt tokens; output assumes generated tokens.
| Volume | Input cost | Output cost | Total (1:1) |
|---|---|---|---|
| 1K tokens | $0.0025 | $0.01 | $0.0125 |
| 1M tokens | $2.50 | $10.00 | $12.50 |
| 10M tokens | $25.00 | $100.00 | $125.00 |
GPT-4o monthly cost calculator
$7.50
per month for 1M in + 500K out
$90.00
per year
Need to weigh it against other models? Use the full cross-provider calculator.
GPT-4o vs Claude Sonnet 4 vs Gemini 2.5 Pro vs DeepSeek-V3
GPT-4o ranks #3 of 4 by blended per-million-token price in this comparison. Lower input + output is cheaper.
| Model | Input $/M | Output $/M | In+Out | Context |
|---|---|---|---|---|
| DeepSeek-V3DeepSeek | $0.27 | $1.10 | $1.37 | 128K |
| Gemini 2.5 ProGoogle | $1.25 | $10.00 | $11.25 | 1M |
| YOUGPT-4oOpenAI | $2.50 | $10.00 | $12.50 | 128K |
| Claude Sonnet 4Anthropic | $3.00 | $15.00 | $18.00 | 200K |
Related pricing pages
GPT-4o pricing FAQ
How much does GPT-4o cost?
GPT-4o is priced at $2.50 per million input tokens and $10.00 per million output tokens on OpenAI. At a typical 2:1 input-to-output ratio, that works out to roughly $5.00 per million blended tokens. Context window is 128K.
Is GPT-4o cheaper than Claude Sonnet 4?
Yes — GPT-4o is cheaper. Its blended input+output rate ($12.50/M) undercuts Claude Sonnet 4 ($18.00/M) by about 31%. On input alone GPT-4o is $2.50/M vs $3.00/M; on output $10.00/M vs $15.00/M.
Does GPT-4o support prompt caching?
Yes. OpenAI serves repeated input prefixes on GPT-4o at $1.25/M — a 50% discount versus the standard $2.50/M input rate. Caching is automatic for qualifying prefixes and is most valuable for long system prompts and recurring context.
Can I use GPT-4o with the Batch API?
Yes. GPT-4o supports the OpenAI Batch API at $1.25/M input and $5.00/M output — 50% off standard rates — in exchange for asynchronous (up to 24-hour) completion. Batch is ideal for offline evaluation, classification, and bulk generation.