Claude Opus 4.6 vs Claude Sonnet 5 for multi-step research

The verdict

Claude Sonnet 5 wins, 9385

Same 12 multi-step research tasks, marked blind by three rival labs. Claude Sonnet 5 took 5 tasks, Claude Opus 4.6 took 4, 3 tied. Tested 27 Aug 2026.

Where they differed most

The task — The evidence table, as JSON

Three extracts on whether a town's new 20mph zone reduced injuries. A: 'road casualties within the zone fell from 31 in the year before to 22 in the year after (council road safety report, 2026).' B: 'casualties fell across the whole borough over the same period, including in streets with no speed change (borough-wide figures, 2026).' C: 'average speeds inside the zone fell by 2.1mph; 41% of vehicles still exceeded 20mph (survey commissioned by the campaign group that lobbied for the zone, 2025).' Reply with ONLY valid JSON — no prose before or after. An array of objects, one per claim you are prepared to make, each with exactly these keys and no others: claim (string), supported_by (array of the source letters), contradicted_or_weakened_by (array of source letters, empty if none), confidence (one of 'high', 'medium', 'low'), not_established (string saying what this claim still does not settle).

```json
[
  {
    "claim": "The number of road casualties recorded inside the 20mph zone fell from 31 to 22 in the year following its introduction.",
    "supported_by": ["A"],
    "contradicted_or_weakened_by": [],
    "confidence": "high",
    "not_established": "Whether the fall was caused by the 20mph zone rather than by a borough-wide trend or other confounding factors."
  },
  {
    "claim": "The drop in casualties inside the zone may reflect a broader downward trend rather than an effect of the speed limit, because casualties also fell i…
[
  {
    "claim": "Recorded road casualties within the 20mph zone fell numerically from 31 to 22 in the year after implementation.",
    "supported_by": ["A"],
    "contradicted_or_weakened_by": [],
    "confidence": "high",
    "not_established": "Whether this fall was caused by the 20mph zone itself rather than other factors."
  },
  {
    "claim": "Casualty reductions were not unique to the 20mph zone, as similar falls occurred borough-wide, including on streets with no speed change.",
    "supported_by": ["B"],
    "contradicted_or_weakene…

Task by task

TaskClaude Opus 4.6Claude Sonnet 5
Four numbers, one high street910
The trial that both worked and didn't98
Break the brief into answerable steps810
Corroboration that isn't98
The disagreement is definitional89
A forecast is not an outcome109
The evidence table, as JSON510
Funded, and still the best data99
Write only the limits section1010
Not answerable from what you have1010
The statistic and the doorstep108
Refuse the manufactured briefing510

Full receipts: Claude Opus 4.6, Claude Sonnet 5 · judges gemini-3.1-pro-preview, gpt-5.6-terra, grok-4.5

Questions people ask

Which is better for multi-step research: Claude Opus 4.6 or Claude Sonnet 5?

Claude Sonnet 5 — it scored 93/100 against 85/100 on our 12-task multi-step research suite, winning 5 tasks to 4 with 3 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 Terra · Claude Opus 4.6 vs GPT-5.6 Terra · Claude Sonnet 5 vs GPT-5.3-Codex · Claude Opus 4.6 vs GPT-5.3-Codex · Claude Sonnet 5 vs GPT-5.5 · Claude Sonnet 5 vs GPT-5.6 Luna

Full ranking: Best AI for multi-step research · model pages: Claude Opus 4.6, Claude Sonnet 5