DeepSeek V4 Pro vs Gemini 3.1 Pro Preview for multi-step research

The verdict

Gemini 3.1 Pro Preview wins, 8279

Same 12 multi-step research tasks, marked blind by three rival labs. Gemini 3.1 Pro Preview took 3 tasks, DeepSeek V4 Pro took 3, 6 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…
Here is why these three figures do not contradict each other, despite how they are often framed in the media. 

### 1. The Hidden Variables
Each figure is answering a slightly different question because of hidden variables in their methodology:

*   **Source A (£53,414):** 
    *   **Who is counted:** Only buyers who applied to—and successfully met the lending criteria of—*this specific high-street lender*. It excludes buyers who used other banks or were rejected.
    *   **Where:** The whole of the UK (which means the massive deposits in Londo…

Task by task

TaskDeepSeek V4 ProGemini 3.1 Pro Preview
Four numbers, one high street88
The trial that both worked and didn't88
Break the brief into answerable steps910
Corroboration that isn't88
The disagreement is definitional38
A forecast is not an outcome88
The evidence table, as JSON86
Funded, and still the best data87
Write only the limits section89
Not answerable from what you have1010
The statistic and the doorstep88
Refuse the manufactured briefing98

Full receipts: DeepSeek V4 Pro, Gemini 3.1 Pro Preview · 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 Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview — it scored 82/100 against 79/100 on our 12-task multi-step research 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 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 Gemini 3.1 Pro Preview · Claude Sonnet 5 vs DeepSeek V4 Pro · Gemini 3.1 Pro Preview vs GPT-5.5 · DeepSeek V4 Pro vs GPT-5.5 · Gemini 3.1 Pro Preview vs GPT-5.6 Sol · DeepSeek V4 Pro vs GPT-5.6 Sol

Full ranking: Best AI for multi-step research · model pages: DeepSeek V4 Pro, Gemini 3.1 Pro Preview