Abacus.AI · LIVEBENCH 2026-06-25

Smaug Agentic

Published benchmark results for this specific model configuration.

smaug-agentic

Overall score79.5 / 100

Cost / successful task$0.329USD · benchmark workload

Weight accessOpen weights

Where this configuration performs

Category averages from the same release, on a 0–100 scale.

Reasoning90.3

Coding82.5

Agentic Coding64.6

Mathematics83.9

Data Analysis79.9

Language84.4

Instruction Following71.0

Task-level results

theory of mind78.8zebra puzzle98.3spatial100.0logic with navigation84.0code generation84.5code completion80.4javascript77.3typescript56.7python60.0AMPS Hard98.0integrals with game52.0math comp94.1olympiad91.6consecutive events90.4tablejoin51.2tablereformat98.0connections100.0plot unscrambling71.1typos82.0paraphrase68.6simplify66.6story generation76.7summarize72.0

Source files: Scores ↗ Categories ↗ Costs ↗ . Missing cost is unknown, never free. Compare configurations within the same release.