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 | OpenAI | Google | Qwen | OpenAI |
|---|---|---|---|---|
| Arena ELOHuman preference | 1,453 | 1,473 | 1,416 | 1,435 |
| LiveBench 1.0General reasoning | 73.6% | Not in our dataset | Not in our dataset | Not in our dataset |
| GPQA DiamondScientific knowledge | Not in our dataset | 90.4% | 86.6% | 85.7% |
| HLEAcademic reasoning | Not in our dataset | 33.7% | 25.3% | 24.8% |
| MMLU ProMulti-task understanding | Not in our dataset | Not in our dataset | 86.7% | 87.1% |
| SimpleBenchCommonsense reasoning | 46.8% | 61.1% | Not in our dataset | 56.7% |
| AIME 2025Mathematics | Not in our dataset | Not in our dataset | Not in our dataset | 94.6% |
| GSOGraduate science | Not in our dataset | 9.8% | Not in our dataset | 6.9% |
| AiderCode editing | Not in our dataset | Not in our dataset | Not in our dataset | 88.0% |
| SWE-Bench VerifiedAgentic coding | Not in our dataset | 78.0% | 72.0% | 72.8% |
| Tau-BenchAgentic tool use | Not in our dataset | Not in our dataset | Not in our dataset | 81.1% |
| Terminal-BenchAgentic terminal | Not in our dataset | 47.6% | Not in our dataset | 35.2% |
| ARC-AGI-2Visual reasoning | 47.6% | 33.6% | Not in our dataset | 9.9% |
Pricing & context Published USD rates per 1M tokens
| Specification | GPT-5.6 Luna Thinking | Gemini 3 Flash 3.0 | 3.5 122B A10B Thinking | GPT-5.0 |
|---|---|---|---|---|
| Input | $0.2 | $0.5 | $0.26 | $1.25 |
| Output | $1.2 | $3 | $2.08 | $10 |
| Context window | 1.1M | Not available | 262K | 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.