OpenAI: o3 Deep Research
★ Elite9.0
/ 10
o3-deep-research is OpenAI's advanced model for deep research, designed to tackle complex, multi-step research tasks. Note: This model always uses the 'web_search' tool which adds additional cost.
Specifications
| Attribute | Value |
|---|---|
| Lab | OpenAI |
| Tags | Reasoning Agentic Intelligent |
| Overall Score | 9.0/10 |
| Release Date | 2025-10 |
| Context Window | 200,000 tokens |
| Input Price / 1M | $10.00 |
| Output Price / 1M | $40.00 |
| Input Modalities | Image, Text, File |
| Output Modalities | Text |
Strengths
- Best-in-class multi-step research and synthesis
- Thorough source evaluation and cross-referencing
- Handles ambiguous open-ended questions well
- Produces cited, well-structured research reports
Weaknesses
- Very expensive at $10/$40 per 1M tokens
- High latency — minutes per query
- Overkill for simple factual lookups
Best For
- Deep research reports and analysis
- Multi-source literature review
- Complex investigative questions
Sources & Further Reading
Related Models
Scores are aggregated from public benchmarks (MMLU, HumanEval, MATH, GSM8K, LMSYS) and normalized to a 1–10 scale. Methodology →