Home Model series Claude Fable Claude Fable 5.2 Claude Fable 5.2 · WANDR · 500-task wide research benchmark · Scores and configurations Claude Fable 5.2 · WANDR · 500-task wide research benchmark · Scores and configurations Claude Fable 5.2 record for WANDR · 500-task wide research benchmark. 5 score records for this release, including 5 exact values (may include historical records). Includes 1 source location.
Scores and evaluation conditions Claude Fable 5.2 · WANDR scores and configurations Metric and score Configuration and evaluation conditions Sources Soft F1 71.4% Reported 71.4% · Unit % Published Unconfirmed
Low effort Settings Claude Fable 5.2;adaptive thinking always on;Low effort。 Unknowns none recorded Tools Tools allowed Reasoning budget low Framework not recorded Protocol WANDR 500-task variant; low effort; web search/fetch; 980k-token budget; soft F1 averaged over tasks. Conditions Frozen search index; Claude Opus 4.6 grader. Tooling, prompts and grader differ from Perplexity’s version. View this score in rankings Soft F1 73.6% Reported 73.6% · Unit % Published Unconfirmed
Medium effort Settings Claude Fable 5.2;adaptive thinking always on;Medium effort。 Unknowns 主文未列该配置的采样设置与运行数。 Tools Tools allowed Reasoning budget medium Framework not recorded Protocol WANDR 500-task variant; medium effort; web search/fetch; 980k-token budget; soft F1 averaged over tasks. Conditions Frozen search index; Claude Opus 4.6 grader. Tooling, prompts and grader differ from Perplexity’s version. View this score in rankings Soft F1 79.3% Reported 79.3% · Unit % Published Unconfirmed
High effort Settings Claude Fable 5.2;adaptive thinking always on;High effort。 Unknowns 主文未列该配置的采样设置与运行数。 Tools Tools allowed Reasoning budget high Framework not recorded Protocol WANDR 500-task variant; high effort; web search/fetch; 980k-token budget; soft F1 averaged over tasks. Conditions Frozen search index; Claude Opus 4.6 grader. Tooling, prompts and grader differ from Perplexity’s version. View this score in rankings Soft F1 80.6% Reported 80.6% · Unit % Published Unconfirmed
Xhigh effort Settings Claude Fable 5.2;adaptive thinking always on;Xhigh effort。 Unknowns 主文未列该配置的采样设置与运行数。 Tools Tools allowed Reasoning budget xhigh Framework not recorded Protocol WANDR 500-task variant; xhigh effort; web search/fetch; 980k-token budget; soft F1 averaged over tasks. Conditions Frozen search index; Claude Opus 4.6 grader. Tooling, prompts and grader differ from Perplexity’s version. View this score in rankings Soft F1 82.1% Reported 82.1% · Unit % Published Unconfirmed
Adaptive Max Settings Claude Fable 5.2;adaptive thinking always on;Max effort;默认 temperature/top_p;Card 通用表说明通常五次运行平均,个别测试另行注明。 Unknowns System Card pp.1-264 文字已完整核读;各benchmark章节未明确披露的运行参数仍未知。 Tools Tools allowed Reasoning budget max Framework not recorded Protocol WANDR 500-task variant; max effort; web search/fetch; 980k-token budget; soft F1 averaged over tasks. Conditions Frozen search index; Claude Opus 4.6 grader. Tooling, prompts and grader differ from Perplexity’s version. View this score in rankings
Benchmark versions and metrics Aliases Wide ANd Deep Research
Soft F1 · % · Higher is better