Compare models. Find the differences.
Add, remove, or change models in the table.
Benchmark snapshot
1 benchmarks in common.
4 models selected. 1 benchmarks in common.
Reported scores; test settings may differ.
| Benchmark | Microsoft | Google | Qwen | Meta |
|---|---|---|---|---|
| Arena ELOHuman preference | Not in our dataset | Not in our dataset | 1,154 | 1,211 |
| LiveBench 1.0General reasoning | 25.3% | Not in our dataset | Not in our dataset | 26.0% |
| GPQA DiamondScientific knowledge | 25.2% | 24.8% | 54.5% | 32.8% |
| MMLU ProMulti-task understanding | 52.8% | 40.5% | Not in our dataset | Not in our dataset |
| AIME 2025Mathematics | 5.3% | 6.7% | Not in our dataset | 2.7% |
| LiveCodeBenchCompetitive coding | 10.4% | 13.2% | Not in our dataset | 10.8% |
Pricing & context Published USD rates per 1M tokens
| Specification | Phi-4 Mini Non-Thinking | Gemma 3n 2B IT Non-thinking | QwQ 32B Preview | Llama 3 8B 3.1 (Non-thinking) |
|---|---|---|---|---|
| Input | $0.08 | $0 | $0.2 | $0.05 |
| Output | $0.35 | $0 | $0.2 | $0.08 |
| Context window | 128K | 8K | 33K | 131K |
| 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.