Meta · LIVEBENCH 2026-06-25

Muse Spark 1.1 xHigh Effort

Published benchmark results for this specific model configuration.

muse-spark-1.1-xhigh

Overall score75.3 / 100

Cost / successful task$0.198USD · benchmark workload

Weight accessNot reported

Where this configuration performs

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

Reasoning87.7

Coding77.2

Agentic Coding58.5

Mathematics87.1

Data Analysis72.5

Language74.3

Instruction Following69.6

Task-level results

theory of mind76.9zebra puzzle96.0spatial100.0logic with navigation78.0code generation76.1code completion78.3javascript77.3typescript43.3python55.0AMPS Hard82.0integrals with game83.0math comp96.1olympiad87.5consecutive events71.1tablejoin46.6tablereformat100.0connections98.0plot unscrambling63.0typos62.0paraphrase69.4simplify65.7story generation69.2summarize74.3

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