Claude Fable 5 and Claude Mythos 5
Open paper
Benchmark
Model
Rank
Results
code-generation-on-swe-bench-pro
Claude Mythos 5
–
Resolved: 80.3
code-generation-on-terminal-bench-2-1
mini-SWE-agent + Claude Mythos 5
–
Accuracy: 88.0
Rank counts only results with a code link.