o3
by OpenAI · openai/o3 · #255 cheapest of 325 paid models
Prices updated Jul 22, 2026, 1:31 AM UTC · refreshed hourly
o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following....
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5Gemini 3.6 Flash
Pricing
Input
$2.00
per 1M tokens
Output
$8.00
per 1M tokens
Blended (3:1)
$3.50
3 input : 1 output
Cache read
$0.50
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for o3 with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $3.50 | $0.212 | — |
| 50% | $2.94 | $0.137 | −35% |
| 90% | $2.49 | $0.077 | −64% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates for this provider’s batch API. Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.0034 | $3.40 |
| Document summary | 8,000 | 1,000 | $0.024 | $24.00 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.076 | $76.00 |
| Long-context analysis | 150,000 | 5,000 | $0.34 | $340.00 |
Performance
Intelligence Index
30.4
AA composite quality
Coding Index
—
Math Index
88.3
Agentic Index
—
Output speed
159 tok/s
median
Time to first token
4.84s
median
Time to first answer
4.84s
after reasoning tokens
Artificial Analysis benchmarks
| Benchmark | Score |
|---|---|
| MMLU-Pro | 85.3% |
| GPQA Diamond | 82.7% |
| Humanity's Last Exam | 20.0% |
| LiveCodeBench | 80.8% |
| SciCode | 41.0% |
| MATH-500 | 99.2% |
| AIME | 90.3% |
| AIME 2025 | 88.3% |
| IFBench | 71.4% |
| AA-LCR (Long Context Reasoning) | 69.3% |
| Terminal-Bench Hard | 37.1% |
| τ²-Bench (Telecom) | 80.7% |
Design Arena Elo
| Category | Elo |
|---|---|
| Website | 1,064 |
| UI components | 1,060 |
| Data viz | 1,200 |
| Game dev | 1,093 |
| Code categories | 1,053 |
Specs
Context window
200K
tokens
Max output
100K
tokens
Input modalities
image, text, file
Tokenizer
GPT