Compare models. Find the differences.
Add, remove, or change models in the table.
Benchmark snapshot
3 benchmarks in common.
4 models selected. 3 benchmarks in common.
Reported scores; test settings may differ.
| Benchmark | Anthropic | Google | DeepSeek | xAI |
|---|---|---|---|---|
| Arena ELOHuman preference | 1,414 | 1,433 | 1,436 | 1,411 |
| LiveBench 1.0General reasoning | 74.6% | Not in our dataset | 74.2% | Not in our dataset |
| GPQA DiamondScientific knowledge | 74.9% | 86.9% | 71.2% | 80.2% |
| HLEAcademic reasoning | 6.7% | 16.0% | 8.1% | Not in our dataset |
| MMLU ProMulti-task understanding | 86.6% | Not in our dataset | 83.0% | Not in our dataset |
| SimpleBenchCommonsense reasoning | 58.8% | Not in our dataset | 61.1% | 36.1% |
| AIME 2025Mathematics | 33.9% | Not in our dataset | Not in our dataset | 77.3% |
| AiderCode editing | 70.7% | Not in our dataset | Not in our dataset | 53.3% |
| SWE-Bench VerifiedAgentic coding | 72.5% | Not in our dataset | 73.7% | Not in our dataset |
| LiveCodeBenchCompetitive coding | 46.9% | 72.0% | 55.2% | 70.6% |
| Tau-BenchAgentic tool use | 81.8% | Not in our dataset | Not in our dataset | Not in our dataset |
| ARC-AGI-2Visual reasoning | 1.3% | Not in our dataset | 2.1% | 0.0% |
Pricing & context Published USD rates per 1M tokens
| Specification | Claude Opus 4 4.0 Non-thinking | Gemini 3.1 Flash Lite Latest | V4 Flash | Grok 3 Thinking |
|---|---|---|---|---|
| Input | $15 | $0.25 | $0.091 | $3 |
| Output | $75 | $1.5 | $0.182 | $15 |
| Context window | 200K | Not available | 1.0M | 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.