Compare models. Find the differences.
Add, remove, or change models in the table.
Benchmark snapshot
2 benchmarks in common.
4 models selected. 2 benchmarks in common.
Reported scores; test settings may differ.
| Benchmark | OpenAI | DeepSeek | OpenAI | Qwen |
|---|---|---|---|---|
| Arena ELOHuman preference | 1,390 | 1,458 | 1,391 | Not in our dataset |
| LiveBench 1.0General reasoning | Not in our dataset | 77.4% | Not in our dataset | 71.3% |
| GPQA DiamondScientific knowledge | 82.3% | 72.9% | 81.4% | 64.0% |
| HLEAcademic reasoning | 16.7% | 7.7% | 18.1% | Not in our dataset |
| MMLU ProMulti-task understanding | 83.7% | 82.9% | Not in our dataset | Not in our dataset |
| SimpleBenchCommonsense reasoning | Not in our dataset | 50.9% | 38.7% | Not in our dataset |
| AIME 2025Mathematics | 91.1% | Not in our dataset | 92.7% | 70.4% |
| GSOGraduate science | Not in our dataset | Not in our dataset | 3.6% | Not in our dataset |
| AiderCode editing | Not in our dataset | Not in our dataset | 72.0% | Not in our dataset |
| SWE-Bench VerifiedAgentic coding | 72.0% | 73.6% | 68.1% | Not in our dataset |
| LiveCodeBenchCompetitive coding | 80.4% | 56.8% | 80.2% | 63.5% |
| Tau-BenchAgentic tool use | Not in our dataset | Not in our dataset | 65.6% | Not in our dataset |
| Terminal-BenchAgentic terminal | 24.0% | Not in our dataset | Not in our dataset | Not in our dataset |
| ARC-AGI-2Visual reasoning | 4.4% | 0.8% | 6.1% | Not in our dataset |
Pricing & context Published USD rates per 1M tokens
| Specification | GPT-5 Mini Thinking (5.0) | V4 Pro | o4 Mini o4-mini | 3 14B Thinking |
|---|---|---|---|---|
| Input | $0.25 | $1.6 | $1.1 | $0.12 |
| Output | $2 | $3.2 | $4.4 | $0.24 |
| Context window | 400K | 1.0M | 200K | 41K |
| 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.