Home Model series Claude Haiku Claude Haiku 4.5 Claude Haiku 4.5 · GPQA Diamond · 198-question Diamond subset · Scores and configurations Claude Haiku 4.5 · GPQA Diamond · 198-question Diamond subset · Scores and configurations Claude Haiku 4.5 record for GPQA Diamond · 198-question Diamond subset. 1 score record for this release, including 1 exact value (may include historical records). Includes 1 source location.
Scores and evaluation conditions Claude Haiku 4.5 · GPQA Diamond scores and configurations Metric and score Configuration and evaluation conditions Sources Accuracy 73% Reported 73.0% · Unit % Published Unconfirmed
发布表报告值 Settings SWE-bench: full 500 problems, 50 trials, no test-time compute, 128K thinking budget, default sampling; Terminal-Bench: Terminus 2, 11 runs split between no thinking and 32K thinking; OSWorld: OSWorld-Verified, 100 steps, 4 runs; GPQA/MMMLU/MMMU follow source table. Unknowns AIME 的 Python 工具使用情况按两个分列配置保存;其余各项的逐行配置以主文脚注为准。 Tools Tool access not recorded Reasoning budget Thinking budget 128K Framework not recorded Protocol GPQA Diamond。 Conditions 按发布主表;具体采样/试验数未在图表逐行列出。 View this score in rankings
Sources and disclosures GPQA Diamond row Main text · Introducing Claude Haiku 4.5 Benchmark versions and metrics Aliases GPQA Diamond、GPQA (diamond)
Accuracy · % · Higher is better