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 66. 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 47.1. The single biggest benchmark swing on the page is SWE-bench Pro, 90 to 49.
GPT-5.3 Codex is also the more expensive model on tokens at $2.50 input / $10.00 output per 1M tokens, versus $0.10 input / $0.30 output per 1M tokens for Step 3.5 Flash. That is roughly 33.3x on output cost alone. GPT-5.3 Codex is the reasoning model in the pair, while Step 3.5 Flash is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. GPT-5.3 Codex gives you the larger context window at 400K, compared with 256K for Step 3.5 Flash.
Pick GPT-5.3 Codex if you want the stronger benchmark profile. Step 3.5 Flash only becomes the better choice if you want the cheaper token bill or you would rather avoid the extra latency and token burn of a reasoning model.
GPT-5.3 Codex
88.1
Step 3.5 Flash
60.2
GPT-5.3 Codex
87.3
Step 3.5 Flash
47.1
GPT-5.3 Codex
91.3
Step 3.5 Flash
66.7
GPT-5.3 Codex
93.7
Step 3.5 Flash
78.3
GPT-5.3 Codex
80.3
Step 3.5 Flash
60.8
GPT-5.3 Codex
93
Step 3.5 Flash
87
GPT-5.3 Codex
92.8
Step 3.5 Flash
82.8
GPT-5.3 Codex
97.7
Step 3.5 Flash
84.5
GPT-5.3 Codex is ahead overall, 89 to 66. The biggest single separator in this matchup is SWE-bench Pro, where the scores are 90 and 49.
GPT-5.3 Codex has the edge for knowledge tasks in this comparison, averaging 80.3 versus 60.8. Inside this category, HLE 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 47.1. 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 84.5. Inside this category, AIME 2023 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 78.3. 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 60.2. Inside this category, OSWorld-Verified is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for multimodal and grounded tasks in this comparison, averaging 91.3 versus 66.7. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for instruction following in this comparison, averaging 93 versus 87. Inside this category, IFEval is the benchmark that creates the most daylight between them.
GPT-5.3 Codex has the edge for multilingual tasks in this comparison, averaging 92.8 versus 82.8. Inside this category, MGSM 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.