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 | Meta | Microsoft | Meta | OpenAI |
|---|---|---|---|---|
| Arena ELOHuman preference | 1,318 | 1,256 | Not in our dataset | 1,338 |
| LiveBench 1.0General reasoning | Not in our dataset | 41.6% | Not in our dataset | Not in our dataset |
| GPQA DiamondScientific knowledge | 50.5% | 56.1% | 49.0% | Not in our dataset |
| MMLU ProMulti-task understanding | 68.9% | Not in our dataset | 73.4% | Not in our dataset |
| SimpleBenchCommonsense reasoning | 19.9% | Not in our dataset | 23.0% | Not in our dataset |
| AIME 2025Mathematics | Not in our dataset | 17.3% | Not in our dataset | Not in our dataset |
| LiveCodeBenchCompetitive coding | Not in our dataset | 25.2% | Not in our dataset | Not in our dataset |
| Terminal-BenchAgentic terminal | Not in our dataset | Not in our dataset | Not in our dataset | 7.9% |
| ARC-AGI-2Visual reasoning | Not in our dataset | Not in our dataset | Not in our dataset | 2.6% |
Pricing & context Published USD rates per 1M tokens
| Specification | Llama 3 70B 3.3 | Phi-4 Non-Thinking | Llama 3 405B 3.1 | GPT-5 Nano Thinking (5.0) |
|---|---|---|---|---|
| Input | $0.1 | $0.07 | $4 | $0.05 |
| Output | $0.32 | $0.14 | $4 | $0.4 |
| Context window | 131K | 16K | 131K | 400K |
| 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.