OpenAI boosts GPT-6 Astra and GPT-6.1 Sol default speed ~50% to 50 tok/s
RexDouglass · x · 2026-10-06
OpenAI's Thibault Sottiaux announced a 50% default speed boost for GPT-6 Astra and GPT-6.1 Sol across subscriptions and all Sign in With ChatGPT partners (OpenCode, Pi, Amp, Devin, ...), reaching roughly 50 tokens/s with no user action needed — live within two hours. Commenters note it's still only about half the speed of Opus 5.5.
Related event: OpenAI Boosts GPT-6 Series Default Inference Speed by ~50%(3 posts)→
More from Models
- Reflection AI open-sources Beam, a 501B-A23B MoE trained from scratch under Apache 2.0 — OfirPress · 2026-10-06
- Qwen3 27B on Unsloth keeps hitting "already searched the knowledge base this turn" limit — AudieMurphy135 · 2026-10-06
- Researcher: Chinese models' social dynamics in Delvetown are understudied and underestimated — lfschiavo · 2026-10-06
- NVIDIA's Nemotron coding model scores 535.4/600 on IOI 2026, beating top human, now on Hugging Face — NVIDIAAI · 2026-10-06
- Abliterated Qwen3.8-Flash GGUF quants land on Hugging Face — SC117 · 2026-10-06
- Several Western open-weight models launching this month, Reflection AI's first to rival top Chinese models — latkins · 2026-10-06