Claude Opus 4.8 vs GPT-5.5 for everyday maths

The verdict

Effectively a dead heat: 10098

Same 12 everyday maths tasks, marked blind by three rival labs. Claude Opus 4.8 took 2 tasks, GPT-5.5 took 0, 10 tied. Tested 14 Aug 2026.

Where they differed most

The task — Estimate honestly

Roughly how many working hours are there in a UK full-time year? Give your estimate, show the 2-3 assumptions it rests on, and say which assumption moves the answer most.

# UK Full-Time Working Hours Estimate

**Estimate: roughly 1,700–1,800 hours per year**

## The calculation

Starting point: 52 weeks × 37.5 hours/week = 1,950 hours (gross)

Then subtract time off:
- Annual leave: ~28 days (statutory minimum, often including bank holidays)
- That's roughly 5.6 weeks off

**52 − 5.6 = ~46.4 working weeks × 37.5 hours ≈ 1,740 hours**

## The 2–3 assumptions this rests on

1. **Weekly hours = 37.5** — assuming a standard "9-to-5 with lunch" (7.5 paid hours/day). Many UK contracts use 35, 37, or 40 hours.

2. **Ho…
Rough estimate: **about 1,700 working hours per year** for a UK full-time employee.

Calculation:

\[
(52 \text{ weeks} - 6.6 \text{ weeks off}) \times 37.5 \text{ hours/week}
= 45.4 \times 37.5
\approx 1,700 \text{ hours}
\]

Assumptions:

1. **Full-time week = 37.5 hours**  
   Common UK office pattern: 7.5 hours/day × 5 days.

2. **Holiday = 25 days annual leave + 8 bank holidays**  
   That is **33 days**, or **6.6 working weeks** off.

3. **No adjustment for sickness, overtime, training, or unpaid leave**  
   This is scheduled working tim…

Task by task

TaskClaude Opus 4.8GPT-5.5
VAT both ways1010
Percentage change vs points1010
Unit price comparison1010
Mortgage overpayment intuition1010
Split a bill with a twist1010
Compound vs simple1010
The discount stack1010
Reverse percentage1010
Rate to reality1010
Average of averages trap1010
Estimate honestly109
Refuse false precision109

Full receipts: Claude Opus 4.8, GPT-5.5 · judges gemini-3.1-pro-preview, gpt-5.6-terra, grok-4.5

Questions people ask

Which is better for everyday maths: Claude Opus 4.8 or GPT-5.5?

Effectively a dead heat: Claude Opus 4.8 edged it 100/100 to 98/100 on our everyday maths suite — too close to matter, so pick on price or the product you already use.

How was this tested?

Both models answered the identical published everyday maths 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 everyday maths head-to-heads: Claude Opus 4.8 vs GPT-5.6 Terra · Claude Opus 4.8 vs GPT-5.6 Sol · Claude Opus 4.8 vs GPT-5.6 Luna · Claude Opus 4.8 vs Grok 4.5 · Claude Opus 4.6 vs Claude Opus 4.8 · Claude Opus 4.8 vs GPT-5.3-Codex

Full ranking: Best AI for everyday maths · model pages: Claude Opus 4.8, GPT-5.5