SWE-Bench Failures: When Coding Agents Spiral into 693 Lines of Hallucinations(surgehq.ai)
surgehq.ai
SWE-Bench Failures: When Coding Agents Spiral into 693 Lines of Hallucinations
https://www.surgehq.ai/blog/when-coding-agents-spiral-into-693-lines-of-hallucinations
1 comments
Is this because GPT-5 hallucinates less in general?