D1 Ultra vs Microsoft Decision 1
AA Decision Index v1.0 AA Decision Index v1.0: Decision models answer factual and multiple-choice questions and classify tool-use and safety cases in the AA composite
D1 Ultra65.2
Microsoft Decision 153.8
AA Decision · Humanity's Last Exam text multiple-choice accuracy AA Decision · HLE text MCQ accuracy: Decision models select answers to text-only Humanity’s Last Exam multiple-choice questions
D1 Ultra15
Microsoft Decision 110.9
AA Decision · Omniscience accuracy AA Decision · Omniscience accuracy: Decision models select correct answers to factual knowledge questions
D1 Ultra31.6
Microsoft Decision 128.2
AA Decision · Omniscience non-hallucination rate AA Decision · Omniscience non-hallucination rate: Decision models answer factual questions, scored on avoiding incorrect assertions
D1 Ultra84.5
Microsoft Decision 139.8
AA Decision · When2Call accuracy AA Decision · When2Call accuracy: Decision models decide when a request requires a tool call
D1 Ultra92.6
Microsoft Decision 181.9
AA Decision · WildGuardTest average F1 AA Decision · WildGuardTest average F1: Decision models classify safety-related inputs, scored by average F1
D1 Ultra95.4
Microsoft Decision 188.2
AA Decision cost per task v1.0 AA Decision cost per task v1.0: Decision models run the AA decision workload, measured by cost per task · lower is better
D1 Ultra0
Microsoft Decision 10
AA Decision time per task v1.0 AA Decision time per task v1.0: Decision models run the AA decision workload, measured by elapsed time per task · lower is better
D1 Ultra93
Microsoft Decision 1207
JevBench v1.6.37 · api composite JevBench v1.6.37 · api composite: Decision models produce typed choices, nulls and scores, evaluated by the stated intelligence, calibration, speed and cost composite
D1 Ultra—
Microsoft Decision 169.1
| Benchmark | D1 Ultra | Microsoft Decision 1 |
|---|---|---|
| AA Decision Index v1.0 | 65.2 | 53.8 |
| AA Decision · Humanity's Last Exam text multiple-choice accuracy | 15 | 10.9 |
| AA Decision · Omniscience accuracy | 31.6 | 28.2 |
| AA Decision · Omniscience non-hallucination rate | 84.5 | 39.8 |
| AA Decision · When2Call accuracy | 92.6 | 81.9 |
| AA Decision · WildGuardTest average F1 | 95.4 | 88.2 |
| AA Decision cost per task v1.0 | 0 | 0 |
| AA Decision time per task v1.0 | 93 | 207 |
| JevBench v1.6.37 · api composite | — | 69.1 |
Specs
| Spec | D1 Ultra | Microsoft Decision 1 |
|---|---|---|
| Input, per 1M | !$0.042 | !$0.042 |
| Context | 33K | 33K |
| Weights | Closed | Closed |
| Released | Oct 2026 | Oct 2026 |
| See images | ✓ | — |
| Read PDFs | — | — |
| Hear audio | — | — |
| Watch video | — | — |
| Call tools | — | — |
| Force a tool call | ? | ? |
| Return strict JSON | Partial | Partial |
| Search the web | ? | ? |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
D1 Ultra → Microsoft Decision 1
No image input
Microsoft Decision 1 → D1 Ultra
Nothing known.