NEW YORK, September 2 — ChatGPT has claimed the top spot in PCMag’s updated ranking of artificial intelligence chatbots, a verdict that carries weight for anyone trying to decide which AI assistant deserves their trust—and their screen time. The evaluation, led by writer Ruben Circelli, put the leading contenders through an unusually broad battery of tests. Response quality, reasoning depth, creative writing chops, memory, customization options, pricing, available models, raw performance, overall capabilities, media generation, and voice features all went under the microscope.
ChatGPT emerged as the best overall pick, though the victory came with notable caveats that everyday users should weigh carefully. Behind the win sits OpenAI’s 5-series of large language models, with GPT-5.6 serving as the latest flagship.
The system offers three distinct variants: Luna for lightweight tasks, Terra for balanced everyday use, and Sol for maximum capability. Users can adjust intelligence levels depending on what they need. For visual work, ChatGPT Images 2.0 handles image generation.
What impressed Circelli most was the breadth of excellence. High-quality responses across the board.
Remarkable complex reasoning that holds up under pressure. Creative writing that outclasses rivals. Memory that retains context across long conversations.
And customization that PCMag called peerless. The deep research feature can generate lengthy reports on demand. Coders get a robust AI programming interface.
For professionals and curious hobbyists alike, these tools promise genuine productivity gains. But the review pulls no punches on limitations.
ChatGPT still stumbles. It occasionally provides incorrect responses. The review explicitly questions whether the paid tiers deliver enough value for the cost.
Speaking of which, the paid plans span a wide range—from a modest monthly fee up to a premium tier that runs into the hundreds of dollars, with varying usage levels and features unlocked at each tier. The free plan exists, yet PCMag found it less generous than what Google Gemini offers at no charge.
Voice chat works on both desktop and mobile, with customizable voice and personality settings. Still, it sounds like AI. The technology has not crossed that uncanny valley into true human warmth.
Integrations present another friction point. ChatGPT’s connections to outside tools and services stay locked inside its own walls. Microsoft Copilot and Google Gemini play more openly with broader ecosystems, a difference that matters for users already embedded in Windows or Google’s universe.
PCMag’s testing carries authority because it began when these tools first appeared and has continued as the landscape has shifted. The publication also maintains related guides on AI search engines and AI image generators for readers wanting deeper dives.
The competitive field remains fierce. Google Gemini took honors for best value. Microsoft Copilot won best for Windows users.
Perplexity claimed best for AI search. Claude earned best ChatGPT alternative.
Grok won best for searching X. Each serves different human needs, different workflows, different loyalties. For the millions still figuring out where AI fits into their work and creativity, this ranking offers a roadmap.
The technology keeps accelerating. What Luna, Terra, and Sol can do today will look quaint tomorrow. Watch whether OpenAI addresses that value question and opens its integrations—or whether rivals exploit those gaps to steal the crown.
