Claude Opus 4.8 vs GPT-5.6 Luna
52 SHARED BENCHMARKSAcross 52 shared benchmarks, Claude Opus 4.8 scores higher on 40 and GPT-5.6 Luna on 12. The widest gap is ARC-AGI-3, where Claude Opus 4.8 scores 1.5 against 0.2. GPT-5.6 Luna is the cheaper of the two on tracked API pricing ($0.20 against $5.00 per million input tokens).
At a glance
Release dates per Artificial Analysis. Prices: Artificial Analysis for Claude Opus 4.8; OpenAI's own price page for GPT-5.6 Luna. ◆ = independently verified score.
Where each leadsby capability area · benchmark wins
Biggest gaps
Claude Opus 4.8 pulls furthest ahead on
- AA-Omniscience · Non-hallucination60.7 vs 7.4
- Terminal-Bench 4.021.7 vs 11.6
- SimpleBench64.8 vs 46.8
GPT-5.6 Luna pulls furthest ahead on
- AA-LCR83.7 vs 77.7
- SWE-bench Verified93 vs 88.6
- HMMT Feb. 202698.5 vs 95.5
Every shared benchmark52 · grouped by area
Reasoning 7
Coding 7
Agentic 6
Factuality 3
Instruction Following 1
Long Context 1
Math 5
Multimodal 2
Other shared benchmarks 20
Each score is the one the model's own page shows — the most authoritative tracked result for that benchmark, on the benchmark's standard methodology (source-attributed). ↓ marks lower-is-better metrics; ◆ an independently verified score. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, openai-official), otherwise the lowest tracked offer.