Compare models. Find the differences.
Add, remove, or change models in the table.
Benchmark snapshot
0 benchmarks in common.
4 models selected. 0 benchmarks in common.
Reported scores; test settings may differ.
| Benchmark | Microsoft | Qwen | Qwen | Qwen |
|---|---|---|---|---|
| Arena ELOHuman preference | Not in our dataset | 1,402 | 1,327 | 1,388 |
| LiveBench 1.0General reasoning | Not in our dataset | 67.6% | 59.4% | Not in our dataset |
| GPQA DiamondScientific knowledge | 45.1% | 62.9% | 54.8% | Not in our dataset |
| MMLU ProMulti-task understanding | Not in our dataset | 77.3% | Not in our dataset | Not in our dataset |
| AIME 2025Mathematics | 33.6% | 24.7% | 21.6% | Not in our dataset |
| GSOGraduate science | Not in our dataset | Not in our dataset | Not in our dataset | 4.9% |
| AiderCode editing | Not in our dataset | 61.8% | Not in our dataset | Not in our dataset |
| SWE-Bench VerifiedAgentic coding | Not in our dataset | 34.4% | Not in our dataset | Not in our dataset |
| LiveCodeBenchCompetitive coding | Not in our dataset | 35.3% | 29.8% | Not in our dataset |
| Tau-BenchAgentic tool use | Not in our dataset | 57.0% | Not in our dataset | Not in our dataset |
| Terminal-BenchAgentic terminal | Not in our dataset | Not in our dataset | Not in our dataset | 23.9% |
Pricing & context Published USD rates per 1M tokens
| Specification | Phi-4 Mini Flash Reasoning | 3 235B A22B Non-Thinking | 3 30B A3B Non-thinking | 3 Coder 480B A35B Instruct Non-thinking |
|---|---|---|---|---|
| Input | Not available | $0.455 | $0.12 | $0.3 |
| Output | Not available | $1.82 | $0.5 | $1 |
| Context window | 128K | 131K | 41K | 262K |
| Pricing details | Explore pricing → | Explore pricing → | Explore pricing → | Explore pricing → |
How to read this comparison
Each column keeps the exact model version you selected. Shared benchmarks are the intersection of their reported results; “all coverage” also shows where evidence is missing. Missing results never count as zero.
These are published benchmark scores, not a controlled head-to-head test. The available evidence does not establish matching evaluation settings, so we show no cross-model winner highlights or deltas. Composite quality scores are omitted because they can summarize different benchmark sets.
Scores and statistics are computed upstream. This page reads a fixed data snapshot; its date describes the snapshot, not the date of every measurement.