GLM 5.2 vs GPT-5.6 Sol for multi-step research
GPT-5.6 Sol wins, 89–78
Same 12 multi-step research tasks, marked blind by three rival labs. GPT-5.6 Sol took 10 tasks, GLM 5.2 took 1, 1 tied. Tested 18 Aug 2026.
Where they differed most
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": "Road casualties within the 20mph zone fell from 31 in the year before to 22 in the year after implementation.",
"supported_by": ["A"],
"contradicted_or_weakened_by": [],
"confidence": "high",
"not_established": "Whether the 20mph zone caused the reduction, or whether other factors were responsible."
},
{
"claim": "The casualty reduction observed within the zone cannot be confidently attributed to the 20mph speed limit, because casualties also fell across the whole borough including on streets…[
{
"claim": "Reported road casualties within the 20mph zone fell from 31 in the year before implementation to 22 in the year after.",
"supported_by": ["A"],
"contradicted_or_weakened_by": [],
"confidence": "high",
"not_established": "This before-and-after comparison does not establish that the 20mph zone caused the reduction, whether the change is statistically significant, or whether reporting and exposure were comparable."
},
{
"claim": "The borough-wide decline, including on streets without a speed change, weak…Task by task
| Task | GLM 5.2 | GPT-5.6 Sol |
|---|---|---|
| Four numbers, one high street | 8 | 10 |
| The trial that both worked and didn't | 8 | 9 |
| Break the brief into answerable steps | 9 | 10 |
| Corroboration that isn't | 8 | 9 |
| The disagreement is definitional | 7 | 9 |
| A forecast is not an outcome | 8 | 9 |
| The evidence table, as JSON | 5 | 8 |
| Funded, and still the best data | 7 | 9 |
| Write only the limits section | 9 | 9 |
| Not answerable from what you have | 8 | 10 |
| The statistic and the doorstep | 8 | 5 |
| Refuse the manufactured briefing | 9 | 10 |
Full receipts: GLM 5.2, 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: GLM 5.2 or GPT-5.6 Sol?
GPT-5.6 Sol — it scored 89/100 against 78/100 on our 12-task multi-step research suite, winning 10 tasks to 1 with 1 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 GLM 5.2 · GPT-5.5 vs GPT-5.6 Sol · GLM 5.2 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: GLM 5.2, GPT-5.6 Sol