Best AI for spreadsheets & Excel / head-to-head

GPT-5.6 Terra vs Qwen3.7 Maxfor spreadsheets & Excel

The verdict

GPT-5.6 Terra wins, 98–95

Same 12 spreadsheets & Excel tasks, marked blind by three rival labs. GPT-5.6 Terra took 3 tasks, Qwen3.7 Max took 1, 8 tied. Tested 13 Aug 2026.

Where they differed most

The task — Clean messy data

These postcodes were typed by hand: 'sw1a1aa', 'SW1A 1AA', ' m1 4wb', 'M14WB', 'ec1v9nr'. Give a single Excel formula that normalises any one of them to standard UK format (uppercase, one space before the final three characters). Assume the raw value is in A2.

GPT-5.6 Terra10/10

```excel
=UPPER(LEFT(SUBSTITUTE(TRIM(A2)," ",""),LEN(SUBSTITUTE(TRIM(A2)," ",""))-3)&" "&RIGHT(SUBSTITUTE(TRIM(A2)," ",""),3))
```

Qwen3.7 Max8/10

## The Formula

```excel
=LET(c, UPPER(SUBSTITUTE(A2," ","")), LEFT(c, LEN(c)-3) & " " & RIGHT(c, 3))
```

## How it works

The formula does three things in sequence:

| Step | Action | Example (`' m1 4wb'`) |
|---|---|---|
| **1. Clean** | `UPPER(SUBSTITUTE(A2," ",""))` — strips *all* spaces and uppercases | `M14WB` |
| **2. Outward** | `LEFT(c, LEN(c)-3)` — grabs everything except the last 3 chars | `M1` |
| **3. Inward** | `RIGHT(c, 3)` — grabs the last 3 chars | `4WB` |

These are then joined with a single space: **`M1 4WB`**

## Verificati…

Task by task

Task	GPT-5.6 Terra	Qwen3.7 Max
Excel formula from a description	10	10
Fix a broken formula	8	7
Clean messy data	10	8
Choose the right chart	10	10
Google Sheets QUERY	10	10
Spot the error in the numbers	10	10
Pivot logic in words	10	10
Percentage change trap	10	10
Conditional formatting rule	10	10
Reconcile two lists	9	10
Explain a formula back	10	10
Refuse the impossible	10	9

Full receipts: GPT-5.6 Terra, Qwen3.7 Max · judges claude-sonnet-5, gemini-3.1-pro-preview, grok-4.5

Questions people ask

Which is better for spreadsheets & Excel: GPT-5.6 Terra or Qwen3.7 Max?

GPT-5.6 Terra — it scored 98/100 against 95/100 on our 12-task spreadsheets & Excel suite, winning 3 tasks to 1 with 8 tied. Every answer was marked blind by three judges from three rival AI labs.

How was this tested?

Both models answered the identical published spreadsheets & Excel 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 spreadsheets & Excel head-to-heads: GPT-5.3-Codex vs GPT-5.6 Terra · GPT-5.3-Codex vs Qwen3.7 Max · GPT-5.6 Luna vs GPT-5.6 Terra · GPT-5.6 Luna vs Qwen3.7 Max · GPT-5.5 vs GPT-5.6 Terra · GPT-5.6 Sol vs GPT-5.6 Terra

Full ranking: Best AI for spreadsheets & Excel · model pages: GPT-5.6 Terra, Qwen3.7 Max