GPT-5.2
Multimodal
#1GPQA Diamond
#1GDPVal
#2SWE-rebench
+2 more
by OpenAI
+
+
+
+
About
GPT-5.2, released by OpenAI in December 2025, was developed as a rapid response to competitive pressure — with improvements concentrated in general intelligence, long-context understanding, agentic tool-calling, and vision. It represented OpenAI's push to maintain momentum during a period of intense model releases from competing labs in the second half of 2025.
+
+
+
+
Timeline
ReleasedDec 11, 2025
Knowledge CutoffAug 1, 2025
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown
Overall Performance
9 benchmarks
Average Score
61.0%
Best Score
93.2%
High Performers (80%+)
2Top Categories
Science
93.2%
Coding
65.2%
Tool Use
60.6%
Finance
59.0%
Agents
49.7%
+
+
+
+
All Benchmark Results for GPT-5.2
Complete list of benchmark scores with detailed information
| GPQA Diamond | Science | 93.20 | 93.2% | Unverified | |
| SWE Bench Verified | Coding | 80.00 | 80.0% | Unverified | |
| Terminal Bench 2.0 | Coding | 64.70 | 64.7% | Unverified | |
| MCP-Atlas | Tool Use | 60.60 | 60.6% | Unverified | |
| Finance Agent | Finance | 59.00 | 59.0% | Unverified | |
| ARC-AGI-2 | Reasoning | 54.20 | 54.2% | Unverified | |
| SWE-rebench | Coding | 51.00 | 51.0% | Unverified | |
| GDPVal | Agents | 49.70 | 49.7% | Unverified | |
| Humanity's Last Exam | Reasoning | 36.60 | 36.6% | Unverified |