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 | Google | OpenAI | MiniMax | Zhipu AI |
|---|---|---|---|---|
| Arena ELOHuman preference | 1,445 | Not in our dataset | 1,391 | 1,441 |
| LiveBench 1.0General reasoning | 82.4% | Not in our dataset | Not in our dataset | Not in our dataset |
| GPQA DiamondScientific knowledge | 86.4% | Not in our dataset | 85.2% | 85.7% |
| HLEAcademic reasoning | 21.6% | Not in our dataset | 19.4% | 24.8% |
| MMLU ProMulti-task understanding | 86.0% | Not in our dataset | Not in our dataset | Not in our dataset |
| SimpleBenchCommonsense reasoning | 62.4% | Not in our dataset | Not in our dataset | 47.7% |
| AIME 2025Mathematics | 88.0% | Not in our dataset | 86.3% | Not in our dataset |
| GSOGraduate science | 3.9% | Not in our dataset | Not in our dataset | Not in our dataset |
| AiderCode editing | 83.1% | 84.9% | Not in our dataset | Not in our dataset |
| SWE-Bench VerifiedAgentic coding | 59.6% | Not in our dataset | Not in our dataset | 73.8% |
| LiveCodeBenchCompetitive coding | 70.4% | Not in our dataset | Not in our dataset | Not in our dataset |
| Tau-BenchAgentic tool use | 67.0% | Not in our dataset | Not in our dataset | Not in our dataset |
| Terminal-BenchAgentic terminal | 32.6% | Not in our dataset | 42.2% | 41.0% |
| ARC-AGI-2Visual reasoning | 4.9% | 4.9% | 4.9% | Not in our dataset |
Pricing & context Published USD rates per 1M tokens
| Specification | Gemini 2.5 Pro | o3 Pro (Extended Reasoning) | M2 M2.5 | GLM-4 4.7 |
|---|---|---|---|---|
| Input | $1.25 | $20 | $0.255 | $0.6 |
| Output | $10 | $80 | $1.02 | $2.2 |
| Context window | Not available | 200K | 205K | 205K |
| 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.