GPT6.1 Sol slashes costs, beats rivals

OpenAI announced GPT-6.1 Sol on September 29 2026 according to The Rundown AI promising near-Astra intelligence for a fifth of the price with new pricing at two dollars per million input tokens and ten dollars output.

Key Takeaways

• GPT-6.1 Sol matches GPT-6 Astra performance on DeepSWE real-world coding benchmarks at roughly one fifth the cost while delivering strong results on AutomationBench and OSWorld 2.0 computer use tasks.
• New Pro 500 plan offers twenty five times usage of Plus tier and Ultrafast mode delivers up to eight times faster speeds in Codex alongside the Decisions API for instant multiple choice routing from Luna.
• Dots AI agents run on GPT-6 Astra connect to over four thousand apps and operate independently in ChatGPT Slack and Teams for continuous workflow automation.

Performance Benchmarks and Technical Details

The model ties GPT-6 Astra on DeepSWE coding tasks at one fifth cost beats Opus 5.5 by 2.2 points on AutomationBench at one third cost and falls only 2.1 points short on OSWorld 2.0 at one seventh cost. These metrics highlight cost efficient scaling for enterprise deployment where high volume token usage drives savings.

Integration with Dots Agents

Dots agents receive their own computer environment and maintain independent operation across thousands of applications. This setup supports long running tasks without constant human oversight and integrates directly into existing collaboration tools.

Business Impact and Opportunities

Companies gain immediate monetization paths through reduced inference expenses allowing wider adoption of AI coding assistants and office automation. Implementation involves starting with Decisions API for routing then scaling to Dots for end to end processes. Challenges include data privacy compliance and agent oversight which are addressed by built in multiple choice controls and usage caps in Pro 500 plans. Key players such as OpenAI position against competitors by emphasizing price performance ratios that open new markets in small and medium businesses previously priced out of frontier models.

Future Outlook

Industry shifts toward agentic systems like Dots will accelerate automation in software development and operations. Regulatory considerations around autonomous agents require transparent logging while ethical best practices emphasize human review loops. Competitive landscape favors providers delivering near frontier performance at lower costs predicting broader market consolidation around efficient models by 2027.

Frequently Asked Questions

What benchmarks does GPT-6.1 Sol lead on?

It ties GPT-6 Astra on DeepSWE and outperforms Opus 5.5 on AutomationBench according to The Rundown AI report.

How does pricing compare to prior models?

Pricing sits at two dollars input and ten dollars output per million tokens enabling five to seven times cost reductions versus comparable frontier systems.

What new plans and APIs were released?

Pro 500 delivers twenty five times usage and Ultrafast speeds while Decisions API supports instant routing for agents and content sorting.

Are Dots agents available now?

Dots agents powered by GPT-6 Astra connect to four thousand plus apps and appear in ChatGPT Slack and Teams following the September 2026 Dev Day announcement.