Hebbia evaluates GPT-6 Astra on financial research tasks
What happened
Between August and September 2026, Hebbia tested a pre-release version of OpenAI's GPT-6 Astra using the same financial research cases and scoring criteria. On November 20, OpenAI News published the results: GPT-6 Astra scored 13.6 percentage points above the best comparison model on citing the required source types, and 99.5% of its citation excerpts were relevant to the question. On presentation decks, the model met 97.9% of requirements, 14.4 points ahead of the best comparison model, and ranked first on both following the brief and slide quality.
Written by AI from the coverage · updated 2 days ago
Coverage
Follow the reports to see the story from different sides.
- OpenAI NewsHebbia sees improvements in financial research with GPT-6 Astra
Hebbia 测试 OpenAI GPT-6 Astra 用于金融研究,在引用所需来源类型上比最佳对比模型高 13.6 个百分点,99.5% 的引用摘录与问题相关。GPT-6 Astra 还满足 97.9% 的演示文稿要求,领先最佳对比模型 14.4 个百分点,并在遵循简报与幻灯片质量两项上均排名第一。Hebbia 于 2026 年 8-9 月测试了预发布版本,使用相同金融研究案例和评分标准。
Heat over time
Heat now 3·Comparable peak 10(Nov 21 06:00)·Comparable change over 24 hours -51%
The trend only compares accounts observed without gaps, so its range may be smaller than the current heat. Hover or tap the chart for each hour; the left and right arrow keys step through it.