---
title: "OpenAI quietly updates its evaluation metrics for GPT-6 Astra, making changes that appear to favor Astra and continuing to revise other metrics after launch (Emily Forlini/Fortune)"
type: signal
url: https://frontiernewswire.com/s/openai-quietly-updates-its-evaluation-metrics-for-gpt-6-astra-making-827c5f5b
published: 2026-09-06T04:40:01.000Z
byline: "Source: TECHMEME"
---

# OpenAI quietly updates its evaluation metrics for GPT-6 Astra, making changes that appear to favor Astra and continuing to revise other metrics after launch (Emily Forlini/Fortune)

Source: TECHMEME · Published 2026-09-06 · Canonical: https://frontiernewswire.com/s/openai-quietly-updates-its-evaluation-metrics-for-gpt-6-astra-making-827c5f5b

Emily Forlini / Fortune : OpenAI quietly updates its evaluation metrics for GPT-6 Astra, making changes that appear to favor Astra and continuing to revise other metrics after launch — OpenAI has changed several evaluation benchmarks for its GPT-6 Astra model since first publishing a blog post announcement mid-afternoon on Sept. 3.

In this story: [Fortune](https://frontiernewswire.com/companies/fortune), [AI benchmarks](https://frontiernewswire.com/topics/ai-benchmarks)

## Sources

- [Read the original at TECHMEME](https://www.techmeme.com/260906/p1#a260906p1)

## Related on Frontier

- [Researchers say LLMs can be good at natively reading and writing neural network weights; GPT-6 Astra edited Qwen3.5-0.8B's weights to make it better at math (Rikiya Takehi)](https://frontiernewswire.com/s/researchers-say-llms-can-be-good-at-natively-reading-and-writing-e4f14b26)
- [AI in the learning era](https://frontiernewswire.com/s/ai-in-the-learning-era-147d7cc1)

---

[More Latest Signal](https://frontiernewswire.com/) · [Signals](https://frontiernewswire.com/verified-reporting)