$2 / $8per 1M tokens in / out
$0.50cache read · −75%
#375on OpenRouter by tokens · 512M / week
Price sheetper 1M tokens
| Input | $2 | list |
| Output | $8 | |
| Cache read | $0.50 | −75% vs input |
| Context · max output | 200K · 100K | tokens |
Where to buy1 provider · per 1M tokens
| Provider | Input | Output | Context |
|---|---|---|---|
| OpenAIlab | $2 | $8 | 200K |
In OpenAI's line-upper 1M in / out
| o4 Mini High | $1.10 / $4.40 | −45% |
| GPT-3.5 Turbo 16k | $3 / $4 | −7% |
| GPT-5.1 (2025-11-13) | $1.25 / $10 | −2% |
| o3 | $2 / $8 | this model |
| GPT-6.1 Sol | $2 / $10 | +14% |
| GPT-4o | $2.50 / $10 | +25% |
| GPT-4o (2024-08-06) | $2.50 / $10 | +25% |
Cheaper for the same workvs o3
| Muse Spark 1.3 max | -43% | score ×2.4 | $1.60 |
| Grok 4.7 xhigh | -14% | score ×2.3 | $3.74 |
| MiMo v2.6 Pro thinkingOpen | -84% | score ×2.3 | $0.13 |
| Qwen3.8 Max (0902) thinking | -14% | score ×2.2 | $5.41 |
| DeepSeek V4.1 Flash 1017 maxOpen | -85% | score ×2.2 | $0.26 |
Pricing questions
How much does o3 cost?›
$2 per 1M input tokens and $8 per 1M output tokens. Cached input is $0.50.
Is o3 free to use?›
There is no free API tier.
How much does prompt caching save?›
Cache reads cost $0.50 per 1M, 75% below the input price.
Is o3 cheaper than o1?›
Yes: $2 / $8 against $15 / $60, 87% less per token.
What is a cheaper alternative to o3?›
Muse Spark 1.3 at max effort costs -43% per task with a score ×2.4 on Humanity's Last Exam.