Introducing GPT-6.1 Sol
What happened
OpenAI announced GPT‑6.1 Sol, an upgrade to GPT‑6 Sol that narrows the gap with GPT‑6 Astra on agentic coding, computer use, and professional tasks while costing about one‑fifth of Astra’s standard token prices. Cached input now costs $0.10 per million tokens, roughly 95% cheaper than standard input pricing and half the cost of GPT‑6 Sol’s cached input. The model shows notable gains on benchmarks such as DeepSWE v1.1, GDP.pdf, AutomationBench, OSWorld 2.0, and Terminal‑Bench Science 0.1, often matching or exceeding Astra’s performance at a fraction of the cost. It also reduces factual errors from 11.4% to 7.7% at low reasoning effort. These improvements make it a more affordable option for building capable AI agents that reuse context across requests.
Why it matters
For everyday developers and businesses, the lower price and higher accuracy mean they can deploy smarter assistants without the high costs previously required for top‑tier models. This could broaden access to advanced AI tools for routine professional workflows.
Reality check
An automated, sceptical read of the sourceIncremental. The release offers cost reductions and modest performance gains rather than a fundamentally new capability.
Is this a big deal?
Get stories like this every morning
The day's AI news in plain English, in your inbox at 8:30 IST. Free.
More from OpenAI
Introducing ChatGPT WalletOpenAI · 13 Nov 2026Shield helps technicians hand off legacy portal work with OpenAIOpenAI · 13 Nov 2026OpenAI launches ChatGPT Wallet, AI agents pay for purchaseOpenAI · 13 Nov 2026OpenAI disbands its robotics team as its general models start running robotsOpenAI · 13 Nov 2026Yang-Mills, OpenAI proof, and the mass gap problemOpenAI · 13 Nov 2026This page was collected from openai.com and summarised automatically by AIPulsee's pipeline. See how it runs.