GPT-o3

In the latest public YZ Index evaluation on 2026-10-26, GPT-o3 scored 68.1 overall (out of 100), ranking #9 among 15 models. The score aggregates four core dimensions — real code sandbox execution, grounding, engineering judgment, and task communication — with 100% rule-based scoring.

GPT-o3 scores 65 on the Execution dimension, ranking #10 of 15 models. This dimension actually runs model-generated programs in an isolated sandbox to verify compilation, runtime correctness, and edge-case handling — no model-as-judge scoring.

In the WDCD instruction-decay test (2026-10-28), which measures instruction compliance under multi-turn pressure, GPT-o3 scored 92.3 (out of 100), ranking #3 of 15 tested models. WDCD applies progressive distraction and social-engineering pressure, with 100% rule-based scoring.

GPT-o3's API is priced at $10 per million input tokens and $40 per million output tokens (official pricing pages are verified regularly). See the Value dimension on this page for price-performance.

The YZ Index runs a full evaluation weekly and sampled evaluations daily; this page updates automatically with every public run. Current data comes from Run #397 on 2026-10-26; see the trend chart for history.