Best AI for summarising documents / head-to-head

Claude Opus 4.8 vs GPT-5.3-Codexfor summarising documents

The verdict

Claude Opus 4.8 wins, 9996

Same 12 summarising documents tasks, marked blind by three rival labs. Claude Opus 4.8 took 3 tasks, GPT-5.3-Codex took 0, 9 tied. Tested 13 Aug 2026.

Where they differed most

The task — Five-bullet summary

Summarise in exactly 5 bullets, max 15 words each, readable in 20 seconds: 'Our Q3 revenue rose 12% to £1.4m, driven mainly by the new retail channel which contributed £310k. Margins slipped from 41% to 37% because of higher shipping costs and a one-off warehouse move. Headcount stayed flat at 34. The main risk into Q4 is the supplier contract, which expires in November and has not been renegotiated. Cash at bank is £620k.'

- Q3 revenue up 12% to £1.4m, boosted by new retail channel.
- Retail channel contributed £310k to overall growth.
- Margins fell from 41% to 37% on shipping and warehouse costs.
- Headcount flat at 34; cash at bank £620k.
- Key Q4 risk: supplier contract expires November, not yet renegotiated.
- Q3 revenue increased 12% to £1.4m.
- New retail channel drove growth, contributing £310k.
- Margins fell from 41% to 37% due shipping costs and warehouse move.
- Headcount remained unchanged at 34 employees.
- Q4 risk: November supplier contract unrenegotiated; cash at bank is £620k.

Task by task

TaskClaude Opus 4.8GPT-5.3-Codex
Meeting notes to actions109
Five-bullet summary108
Do not invent1010
One-sentence gist1010
Preserve the caveat1010
Summarise a disagreement109
Extract the decision1010
Long to short, no loss1010
Summary for a specific reader99
Bullet the risks only1010
Headline and standfirst1010
Say when there is nothing to summarise1010

Full receipts: Claude Opus 4.8, GPT-5.3-Codex · judges gemini-3.1-pro-preview, gpt-5.6-terra, grok-4.5

Questions people ask

Which is better for summarising documents: Claude Opus 4.8 or GPT-5.3-Codex?

Claude Opus 4.8 — it scored 99/100 against 96/100 on our 12-task summarising documents suite, winning 3 tasks to 0 with 9 tied. Every answer was marked blind by three judges from three rival AI labs.

How was this tested?

Both models answered the identical published summarising documents 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 summarising documents head-to-heads: Claude Opus 4.8 vs GPT-5.5 · Claude Opus 4.8 vs GPT-5.6 Luna · Claude Opus 4.8 vs GPT-5.6 Terra · Claude Opus 4.8 vs DeepSeek V4 Pro · Claude Opus 4.8 vs GPT-5.6 Sol · Claude Fable 5 vs Claude Opus 4.8

Full ranking: Best AI for summarising documents · model pages: Claude Opus 4.8, GPT-5.3-Codex