DeepSeek V4 Pro 0813 outpaces GPT-5.6 Sol and Claude Fable 5 on DeepSWE
DeepSeek V4 Pro 0813 hits 88.5% pass@4 on DeepSWE, beating GPT-5.6 Sol (85.8%) and Claude Fable 5 (84.1%) at $0.24 per rollout vs $8.37 and $21.63. A Pro-first cascade escalating to Sol solves 83.0% of tasks at $3.35 each.
2 sources
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Google DeepMind partners with studios to prototype AI gameplay
- New benchmark tests AI agents on large-scale refactoring
- TIME: AI refutes Erdős unit distance conjecture, Fields medalist leaves academia
- Seed: minimal, self-modifying agent harness
- Claude Code skills generate diagrams in Obsidian