Verdict
GPT-5.5 leads 15–5 across 20 shared benchmarks.
Claude Opus 4.8 5 · GPT-5.5 15 · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Reasoning: GPT-5.5 leads 5–0 across 5 shared reasoning benchmarks (largest gap: SimpleQA Verified, 63.1 vs 39.5 — single evaluator).
Agentic: Claude Opus 4.8 leads 2–0 across 2 shared agentic benchmarks (largest gap: OSWorld 2.0, 20.6 vs 13 — single evaluator).
Coding: Evenly split 1–1 across 2 shared coding benchmarks (largest gap: FrontierCode, 46.5 vs 43 — single evaluator).
Math: GPT-5.5 leads 2–0 across 2 shared math benchmarks (largest gap: FrontierMath-Tier-4-v2-Private, 72.5 vs 56.1 — single evaluator).
Unknown: GPT-5.5 leads 2–0 across 2 shared unknown benchmarks (largest gap: EBR-bench, 26.98 vs 21.43 — single evaluator).
Vendor claims — compare against the measured scores above. A claim is what a developer says about its own model, not an independent measurement.
No vendor claims on file for Claude Opus 4.8.
Agentic / Tool Use
“The gains are especially strong in agentic coding, computer use, knowledge work, and early scientific research”
Coding
“It excels at writing and debugging code, researching online, analyzing data, creating documents and spreadsheets, operating software, and moving across tools until a task is finished.”
General
“We're releasing GPT-5.5, our smartest and most intuitive to use model yet, and the next step toward a new way of getting work done on a computer.”
Speed / Latency
“GPT-5.5 matches GPT-5.4 per-token latency in real-world serving, while performing at a much higher level of intelligence.”
GPT-5.5 leads 15–5 across 20 shared benchmarks. Claude Opus 4.8 leads 5 benchmarks and GPT-5.5 leads 15. Higher isn't always better — see the integrity caveats on each benchmark.
Claude Opus 4.8 and GPT-5.5 have 20 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
GPT-5.5 leads reasoning 5–0. Claude Opus 4.8 leads agentic 2–0.