OpenAI says GPT-6 Astra low setting outperforms GPT-5.6 Sol high setting
OpenAI’s Head of Product, Thibaud Sottiaux, announced on X on September 6, 2026 that the new GPT‑6 Astra model’s “low” inference setting delivers higher quality and speed than the “high” setting of the previous‑generation GPT‑5.6 Sol. The official guidance recommends users of Sol who relied on the high setting switch to low or medium modes on Astra.
Key points
- GPT‑6 Astra’s low setting outperforms GPT‑5.6 Sol’s high setting, according to OpenAI.
- Token consumption for long‑tail tasks is reduced by up to 3‑4 times for subscription users.
- OpenAI advises users to downgrade from Sol’s high setting to Astra’s low or medium mode.
OpenAI also said the model includes an optimization that cuts token consumption for long‑tail tasks by up to 3‑4 times for subscription‑plan users, without sacrificing response quality. The reduction is intended to lower inference costs for power users and improve overall API and chat‑based usage efficiency.
Model page: GPT-5.6 Sol →
The story so far
5 episodes →- OpenAI says GPT-6 Astra low setting outperforms GPT-5.6 Sol high settingthis story
[AI] GPT-6 Astra's "low" setting outperforms GPT-5.6 Sol's "high" setting: OpenAI official reveals leap in inference efficiency and token consumption reduction
note.com · 20 October 2026
Loading the full article…
This text was published by note.com and written by 今岡陵. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- Recursive releases Flint 40B coding agent model on Hugging Face · 5 src
- Anthropic, Google and OpenAI prepare new AI models, report says · 1 src
- Ggml-org adds FlintForCausalLM model to llama.cpp · 1 src
- OpenAI says its AI agents weakened internal misalignment monitoring system · 26 src
- Meta launches Muse frontier model family · 1 src
Comments
via GitHub Discussions