Model result, rank 7 of 27

Claude Fable 5

OpenRouter · extra-high reasoning · Full + SWE + Hard measured. Public result card with the model’s overall score, lane measurements, runtime and cost telemetry, and the ranking formula.

Compare all models Download public JSON

Figure 1 Lane profile against the cohort Each bar is one measured lane on the shared 0 to 100 scale. The thin mark is the cohort median; the bright mark is the best score any ranked model reached on that lane.

Full / Agenticlane rank #22 of 27

72.05

SWE MVPlane rank #9 of 27

81.42

Hard Intelligencelane rank #5 of 27

92.78

Claude Fable 5
Cohort median
Cohort best

Overall score82.08

Rank #7

Full / Agentic72.05

Full rank #22

SWE MVP81.42

SWE rank #9

Hard Intelligence92.78

Hard rank #5

Measured cost$12.285

100.0% reliability

Overall

All-around publication view

Score82.08

Formulamean(Full, SWE, Hard Intelligence)

BasisOpenRouter · extra-high reasoning · Full + SWE + Hard measured

The overall score averages the measured major lanes while keeping each source measurement visible.

Lane 01

Full / Agentic benchmark

Final72.05

Capability90.50

Agentic83.95

Pass rate88.6%

Prompts43

This lane captures instruction following, structured behavior, tool discipline, and general agentic reliability.

Lane 02

Software engineering MVP

SWE score81.42

Focused final76.33

Capability79.17

Daily driver75.36

Prompts24

This lane is closer to implementation usefulness: source handling, architecture cleanliness, and deliverable quality.

Lane 03

Hard Intelligence diagnostic

Hard score92.78

Active inquiry99.75

Online adaptation98.13

Self-repair85.73

Authority integrity87.50

Hard Intelligence measures active inquiry, online adaptation, evidence-driven self-repair, and authority/salience integrity.

Telemetry

Runtime economics

Total cost$12.285

Cost / scored item$0.148

Seconds / timed item12.37s

Runtime coverage100.0%

Recorded tokens / item13.1k

Token coverage100.0%

Cost, time, and token basis are normalized telemetry. They explain tradeoffs; they do not overwrite the capability score yet.

Interpretation

Why the result lands here.

The model is stronger in the SWE lane than in the Full/Agentic lane; the overall score is therefore shown with both component lanes visible. Hard Intelligence score is 92.78 and contributes to the overall score alongside Full/Agentic and SWE. Exact OpenRouter model: anthropic/claude-fable-5. Measured with OpenRouter extra-high reasoning across Full/Agentic, SWE MVP, and Hard Intelligence. Hard Intelligence is a public diagnostic, not a hidden official score. Full long-context probe rescored under suite v4 (ambiguous probe: provider content filter returned empty).