AnalysisAI Models
12 hours ago
OpenMythos benchmarks compare with Qwen 3.6 27B
OpenMythos benchmarks released on Reddit compare performance with Qwen 3.6 27B. Qwen's official numbers used a different eval harness and refined/filtered benchmark problems, causing discrepancies. The post details SWE-bench and other eval results.
·
12 hours ago
