GPT-5.2 Codex
GPT-5.2 Codex is a code-tuned variant of GPT-5.2 built for software development. With a 272,000-token context window and tool use, it handles complex coding scenarios such as cross-file consistency, large-scale refactoring, and architectural decisions. It supports structured outputs and is aimed at development teams working through challenging engineering problems where reasoning about code structure and its consequences matters.
Key info
Input
Output
Features
Context window
272K
Max output
128K
Benchmarks
Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.
Global rank#72 of 597 LLMs
TierStrong
Intelligence Index41.2
Reasoning & knowledge
GPQA Diamond
90%
Humanity's Last Exam
36%
Long-context reasoning
79%
Coding
SciCode
55%
Agentic & tool use
Terminal-Bench Hard
37%
τ²-Bench Telecom
92%
Math & instruction following
IFBench
78%
See full leaderboard →Benchmarks via Artificial Analysis · View on AA
Available routes
No routes currently available — GPT-5.2 Codex isn't routed through the Opper gateway right now. It may return.
Contact us about this model →