Introducing Muse Spark 1.1

Open paper
Benchmark Model Rank Results
code-generation-on-swe-bench-proMuse Spark 1.1–Resolved: 61.50
code-generation-on-terminal-bench-2-1bash-tool-only agent harness + Muse Spark 1.1–Accuracy: 80.0

Rank counts only results with a code link.