In the news
Hebbia sees improvements in financial research with GPT-6 Astra
OpenAI · Published · 3 min read
In 30 seconds
- What happened
- Hebbia's Matrix financial research tool showed improvements using GPT-6 Astra, scoring 13.6 percentage points higher on source citation and 97.9% on presentation requirements.
- Why it matters
- Investment bankers and financial analysts need this when preparing acquisition diligence summaries and valuation decks from large document sets.
- Watch out
- Hebbia tested a prerelease build in August-September 2026 using their own grading criteria; independent verification of these results is not yet available.
- rag
- eval
- gpt
The patterns behind this
Each one covers how the technique works, when it earns its cost, and where it breaks.
The Agent Architect
One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.
Weekly email, one-click unsubscribe. We only use your address to send the briefing.