Databricks says Claude Opus 5 tops its coding benchmark and ships in AI Gateway
thesaraharminta · x · 2026-07-25
Databricks says Claude Opus 5 is now available in its AI Gateway, with governed access to data already in the Lakehouse.
An internal Databricks evaluation claims Opus 5 is a clear step up from Opus 4.8:
- Coding: top score on their internal coding bench, beating even Fable 5 while costing about the same as Opus 4.8.
- OfficeQA Pro: 60.9 vs 48.12 for Opus 4.8, and roughly on par with Fable 5.
- Enterprise agentic search: ranked #4 on the bench, but still clearly ahead of Opus 4.8.
Databricks says the model brings better agentic coding, knowledge-work performance, and long-horizon reasoning, with stronger performance per token and lower cost at production scale.
More from Companies & People
- RSNA 2026 Deep Learning Labs return: eight hands-on AI classes over four days in Chicago — PRouzrokh · 2026-10-27
- Revolut founder Storonsky: investor pushed me to fire my best engineer, so I removed all VCs from the board — ctjlewis · 2026-10-27
- Embodied AI startups need a different operational discipline, investor says — ___Mufasaa · 2026-10-27
- Reddit asks: How dependent are Chinese AI labs on the US ecosystem? — lol2funneeee · 2026-10-27
- Stanford HAI's Human-Centered AI for Social Impact program wraps third cohort — StanfordHAI · 2026-10-27
- Goodfire commits $10M to a 'Human Genome Project for AI' to reverse-engineer frontier models — CatAstro_Piyush · 2026-10-27