Abacus.AI · LIVEBENCH 2026-06-25

Smaug Flash

Published benchmark results for this specific model configuration.

smaug-flash

Overall score77.4 / 100

Cost / successful task$0.014USD · benchmark workload

Weight accessOpen weights

Where this configuration performs

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

Reasoning86.2

Coding77.1

Agentic Coding61.1

Mathematics90.1

Data Analysis79.1

Language79.2

Instruction Following69.3

Task-level results

theory of mind76.9zebra puzzle98.0spatial98.0logic with navigation72.0code generation80.3code completion73.9javascript68.2typescript50.0python65.0AMPS Hard97.0integrals with game76.0math comp97.1olympiad90.2consecutive events88.6tablejoin50.7tablereformat98.0connections99.3plot unscrambling60.3typos78.0paraphrase65.6simplify63.1story generation70.0summarize78.3

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