DeepSeek V4 Pro vs GPT-5.6 Sol for multi-step research

The verdict

GPT-5.6 Sol wins, 8979

Same 12 multi-step research tasks, marked blind by three rival labs. GPT-5.6 Sol took 9 tasks, DeepSeek V4 Pro took 1, 2 tied. Tested 18 Aug 2026.

Where they differed most

The task — The disagreement is definitional

Three sources on the average deposit a UK first-time buyer needs. Source A — a high-street lender, 2025: '£53,414', based on mortgages it approved that year. Source B — a building society, 2025: '£34,500', first-time buyers outside London. Source C — a London estate agency, 2024: '£71,000'. These are usually written up as a contradiction. Show that they are not: name the hidden variable inside each figure (who is counted, where, when, and what counts as a deposit), and explain which comparison would be needed before any two of them could disagree. Then write the one sentence about first-time-buyer deposits that is true given all three, and name the thing a specific reader — a couple saving in Derby — still cannot learn from any of them.

They are not contradictions; they are averages of different groups, places, times and definitions.

| Source | Who is counted | Where | When | What counts as a deposit |
|---|---|---|---|---|
| **A: High-street lender, £53,414** | First-time buyers approved for mortgages by that specific lender | UK-wide/lender’s national mortgage book, including London | 2025 | Mortgage deposit at completion: purchase price minus mortgage; may include gifted funds; not a minimum required |
| **B: Building society, £34,500** | First-time buyer mortgage customer…
“Average deposit” is not a single, universal quantity. Each figure is a conditional average:

| Source | Who is counted | Where | When | What “deposit” means |
|---|---|---|---|---|
| **A: £53,414** | First-time buyers whose applications were approved by that particular lender—not rejected applicants, cash buyers or customers of other lenders | The lender’s own geographic mix, which need not represent the UK evenly | Approvals recorded in 2025, rather than necessarily purchases completed in 2025 | Usually the proposed purchase price minus the a…

Task by task

TaskDeepSeek V4 ProGPT-5.6 Sol
Four numbers, one high street810
The trial that both worked and didn't89
Break the brief into answerable steps910
Corroboration that isn't89
The disagreement is definitional39
A forecast is not an outcome89
The evidence table, as JSON88
Funded, and still the best data89
Write only the limits section89
Not answerable from what you have1010
The statistic and the doorstep85
Refuse the manufactured briefing910

Full receipts: DeepSeek V4 Pro, GPT-5.6 Sol · judges claude-sonnet-5, gemini-3.1-pro-preview, gpt-5.6-terra

Questions people ask

Which is better for multi-step research: DeepSeek V4 Pro or GPT-5.6 Sol?

GPT-5.6 Sol — it scored 89/100 against 79/100 on our 12-task multi-step research suite, winning 9 tasks to 1 with 2 tied. Every answer was marked blind by three judges from three rival AI labs.

How was this tested?

Both models answered the identical published multi-step research 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 multi-step research head-to-heads: Claude Sonnet 5 vs GPT-5.6 Sol · Claude Sonnet 5 vs DeepSeek V4 Pro · GPT-5.5 vs GPT-5.6 Sol · DeepSeek V4 Pro vs GPT-5.5 · GPT-5.6 Sol vs Kimi K3 · GPT-5.6 Sol vs Grok 4.5

Full ranking: Best AI for multi-step research · model pages: DeepSeek V4 Pro, GPT-5.6 Sol