Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
GPT-5.3 Codex is clearly ahead on the aggregate, 89 to 85. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GPT-5.3 Codex's sharpest advantage is in coding, where it averages 87.3 against 75.5. The single biggest benchmark swing on the page is SWE-bench Pro, 90 to 77. GPT-5.2 Instant does hit back in instruction following, so the answer changes if that is the part of the workload you care about most.
GPT-5.3 Codex is also the more expensive model on tokens at $2.50 input / $10.00 output per 1M tokens, versus $1.50 input / $6.00 output per 1M tokens for GPT-5.2 Instant. GPT-5.3 Codex gives you the larger context window at 400K, compared with 128K for GPT-5.2 Instant.
Pick GPT-5.3 Codex if you want the stronger benchmark profile. GPT-5.2 Instant only becomes the better choice if instruction following is the priority or you want the cheaper token bill.
GPT-5.3 Codex
88.1
GPT-5.2 Instant
79.6
GPT-5.3 Codex
87.3
GPT-5.2 Instant
75.5
GPT-5.3 Codex
91.3
GPT-5.2 Instant
93.1
GPT-5.3 Codex
93.7
GPT-5.2 Instant
90.9
GPT-5.3 Codex
80.3
GPT-5.2 Instant
79.8
GPT-5.3 Codex
93
GPT-5.2 Instant
95
GPT-5.3 Codex
92.8
GPT-5.2 Instant
94.4
GPT-5.3 Codex
97.7
GPT-5.2 Instant
97.2
GPT-5.3 Codex is ahead overall, 89 to 85. The biggest single separator in this matchup is SWE-bench Pro, where the scores are 90 and 77.
GPT-5.3 Codex has the edge for knowledge tasks in this comparison, averaging 80.3 versus 79.8. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for coding in this comparison, averaging 87.3 versus 75.5. Inside this category, SWE-bench Pro is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for math in this comparison, averaging 97.7 versus 97.2. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for reasoning in this comparison, averaging 93.7 versus 90.9. Inside this category, MRCRv2 is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for agentic tasks in this comparison, averaging 88.1 versus 79.6. Inside this category, OSWorld-Verified is the benchmark that creates the most daylight between them.
GPT-5.2 Instant has the edge for multimodal and grounded tasks in this comparison, averaging 93.1 versus 91.3. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
GPT-5.2 Instant has the edge for instruction following in this comparison, averaging 95 versus 93. Inside this category, IFEval is the benchmark that creates the most daylight between them.
GPT-5.2 Instant has the edge for multilingual tasks in this comparison, averaging 94.4 versus 92.8. Inside this category, MMLU-ProX is the benchmark that creates the most daylight between them.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.