Best AI for customer service replies / head-to-head

Claude Opus 4.6 vs Gemini 3.1 Pro Previewfor customer service replies

The verdict

Gemini 3.1 Pro Preview wins, 97–94

Same 12 customer service replies tasks, marked blind by three rival labs. Gemini 3.1 Pro Preview took 3 tasks, Claude Opus 4.6 took 3, 6 tied. Tested 14 Aug 2026.

Where they differed most

The task — The wrong customer

A furious email arrived meant for a different company ('your driver hit my fence' — you sell software). Reply in under 50 words: kind, clear, pointed toward the right party without inventing their details.

Claude Opus 4.65/10

Subject: Re: Your Driver Hit My Fence

Hi there,

I'm so sorry about your fence — that sounds really frustrating. However, we're a software company, so this message likely reached us by mistake. I'd recommend checking the original company's contact details (perhaps on a receipt or invoice) to get this resolved quickly.

Wishing you luck!

Gemini 3.1 Pro Preview10/10

Hi there, 

I’m so sorry about your fence! You've actually reached the wrong company. We sell software and don't have any drivers or trucks. 

Please double-check the contact info for the delivery service so they can fix this for you. Wishing you a quick resolution!

Task by task

Task	Claude Opus 4.6	Gemini 3.1 Pro Preview
Refund the easy way	10	10
Say no without weasel words	10	10
Reply to a 1-star review	10	10
The confused customer	10	9
De-escalate a threat	10	10
The feature you won't build	9	10
Apologise for your mistake	10	8
Price rise notice	10	10
Outage update mid-incident	10	9
The wrong customer	5	10
Chargeback rebuttal facts	10	10
Refuse the fake review request	9	10

Full receipts: Claude Opus 4.6, Gemini 3.1 Pro Preview · judges gemini-3.1-pro-preview, gpt-5.6-terra, grok-4.5

Questions people ask

Which is better for customer service replies: Claude Opus 4.6 or Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview — it scored 97/100 against 94/100 on our 12-task customer service replies suite, winning 3 tasks to 3 with 6 tied. Every answer was marked blind by three judges from three rival AI labs.

How was this tested?

Both models answered the identical published customer service replies tasks. Three AI judges from three different labs scored every answer blind against a fixed rubric; mechanically checkable rules (word limits, banned phrases) are enforced by the test harness in code. The raw outputs and judge verdicts are downloadable.

More customer service replies head-to-heads: Gemini 3.1 Pro Preview vs GPT-5.6 Sol · Claude Opus 4.6 vs GPT-5.6 Sol · Gemini 3.1 Pro Preview vs GPT-5.5 · Claude Opus 4.6 vs GPT-5.5 · Gemini 3.1 Pro Preview vs GPT-5.3-Codex · Claude Opus 4.6 vs GPT-5.3-Codex

Full ranking: Best AI for customer service replies · model pages: Claude Opus 4.6, Gemini 3.1 Pro Preview