OpenAI's GPT-6.1 Sol gives teams a lower-cost way to handle coding, computer control, and multi-step office tasks. Inside ChatGPT Work and Codex, it lets users write software, read documents, and delegate more involved jobs without stepping up to GPT-6 Astra.
Cost is the hook, with token pricing set at one-fifth of GPT-6 Astra. It also cuts factual errors on hard prompts from 11.4% to 7.7% at low reasoning effort and is more reliable about user limits. Across reasoning settings, its error rate stays within 1.9% of GPT-6 Astra.
For buyers, that creates room to use agentic AI more often in everyday work instead of saving it for the highest-stakes tasks. Competing model providers now have to match that balance of near-flagship capability, lower operating cost, and tighter behavior around user intent.
Image Credit: SeongJoon Cho/Bloomberg / Getty Images
What Makes This Trend Stand Out
- Budget Agentic AI
- Lower-cost autonomous models are broadening access to complex task delegation, creating space for businesses to embed AI agents into routine workflows rather than reserving them for premium use cases.
- Near-flagship Automation
- AI systems that approach top-tier performance at reduced prices are reshaping procurement expectations around capability, reliability, and operating cost.
- Intent-aligned Assistants
- Improved adherence to user limits is elevating trust in delegated digital work, especially for teams using AI across coding, document review, and office execution.
Sectors Adopting This
- Enterprise Software
- Product suites can integrate affordable agentic capabilities into everyday tools, shifting value from standalone assistants toward embedded workflow automation.
- Software Development
- Coding teams gain more scalable access to AI-assisted programming, testing, and debugging as lower-cost models make continuous support economically practical.
- Business Process Automation
- Office operations are becoming more adaptable as multi-step AI agents handle document-heavy and computer-control tasks with fewer accuracy and compliance tradeoffs.
