Now Tracking Claude Sonnet 5
Centium now tracks Claude Sonnet 5. Its reliable knowledge cutoff moved from August 2025 to January 2026, putting more recent information behind the answers we measure.
a newer Claude, now in your tracking.
Centium tracks five AI models: ChatGPT, Claude, Gemini, Perplexity, and Grok. Claude now runs on Anthropic's latest model, Claude Sonnet 5, replacing Claude Sonnet 4.6.
When the company behind a model ships a stronger version, what that model says about brands and categories moves with it. Keeping the model current is what makes the Claude column in your dashboard reflect the answers people get today, not the answers an older version gave.
closer to Opus, at Sonnet speed.
Sonnet 5 is a real step up from Sonnet 4.6, not a routine refresh. Anthropic positions it as approaching the quality of Opus 4.8, their most capable model, while staying in the faster Sonnet tier. The Claude column in your dashboard is now backed by noticeably stronger reasoning at the same point in the lineup.
Sonnet is also the default model on Claude. When someone asks Claude a general question, the answer almost always comes from Sonnet, with Opus held back for the heaviest reasoning. That is the main reason Centium tracks Sonnet rather than Opus: it is the faster model, the better fit for the volume of prompts in a study, and the version the large majority of people actually get. Your data reflects the everyday Claude experience, not a tier most users rarely reach for.
Three of the gains matter most for what Centium measures.
More recent knowledge. The reliable knowledge cutoff moved from August 2025 to January 2026, putting more recent information in the model's training data. Its read on your market, your competitors, and what is current is fresher.
Fewer hallucinations. Anthropic reports lower hallucination rates than Sonnet 4.6. When the model invents less, the brands, claims, and reasons it surfaces about you sit closer to what it actually knows.
Less sycophancy. It is also less inclined to tell a user what they want to hear. Its recommendations and assessments come out more candid, which is the unvarnished read Centium exists to capture.
Beyond those, it reasons more thoroughly before answering and is better at checking its own output, which tends to produce more considered, better-grounded recommendations. Anthropic covers the full release in their announcement.
what to expect in your results.
Your prompts, your categories, and your methodology are unchanged. The only difference is the model answering Claude's share of them.
Expect the first update after the switch to show some movement in the Claude numbers that comes from the stronger model rather than a real change in your market. After that first run, your Claude trend lines are comparable again. The other four models are unaffected.
on your next update.
The upgrade is automatic, with nothing to turn on. Brands processed since the switch already reflect Sonnet 5, and everyone else picks it up on their next scheduled update.
You can confirm which model produced your current data in the Methodology section of your dashboard, which lists every model alongside its knowledge cutoff.
where each model's knowledge lands.
Your dashboard reports each model's training data cutoff, the most recent date in its training set. The reliable knowledge cutoff is the more conservative figure: the point through which that knowledge is most complete and dependable. Sonnet 5 is the most recent Sonnet model on both, and here is how the full line compares.
- Sonnet 3.5
- Reliable Knowledge
- Apr 2024
- Training Data
- Apr 2024
- Sonnet 3.7
- Reliable Knowledge
- Oct 2024
- Training Data
- Nov 2024
- Sonnet 4
- Reliable Knowledge
- Mar 2025
- Training Data
- Mar 2025
- Sonnet 4.5
- Reliable Knowledge
- Jan 2025
- Training Data
- Jul 2025
- Sonnet 4.6
- Reliable Knowledge
- Aug 2025
- Training Data
- Jan 2026
- Sonnet 5Now tracking
- Reliable Knowledge
- Jan 2026
- Training Data
- Jan 2026
questions, answered.
No. The upgrade is automatic. Your next scheduled update runs on the new model, and brands updated since the switch already reflect it.
The first update after the switch can show some movement in the Claude numbers that comes from the stronger model rather than a real change in your market. After that, runs are comparable again. Your prompts and methodology are unchanged.
The Methodology section of your dashboard lists each model and its knowledge cutoff. Claude reads Claude Sonnet 5 once your data has run on it.
More guides
How PR and Earned Media Drive AI Visibility
PR used to be a long game you could not measure. Now earned media is a performance channel, and Centium ties your coverage to AI citations and visibility.
What Content AI Wants From Your Brand
AI does not buy the brand story. It weighs reviews, rankings, specs, and accolades. Here is the content that makes AI confident enough to recommend you.
How to Measure Your Brand's AI Visibility
AI search has no rankings and no analytics. Here is how Centium measures whether AI recommends your brand, across five models and hundreds of prompts.