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 | DeepSeek | xAI | Qwen | OpenAI |
|---|---|---|---|---|
| Arena ELOHuman preference | 1,439 | 1,455 | 1,474 | 1,475 |
| LiveBench 1.0General reasoning | Not in our dataset | 78.0% | 73.1% | 78.0% |
| GPQA DiamondScientific knowledge | 88.1% | Not in our dataset | 90.0% | 92.8% |
| HLEAcademic reasoning | 34.8% | Not in our dataset | Not in our dataset | 43.9% |
| MMLU ProMulti-task understanding | 86.2% | Not in our dataset | Not in our dataset | Not in our dataset |
| SimpleBenchCommonsense reasoning | Not in our dataset | 75.9% | 70.4% | Not in our dataset |
| GSOGraduate science | Not in our dataset | Not in our dataset | Not in our dataset | 31.4% |
| SWE-Bench VerifiedAgentic coding | 79.0% | Not in our dataset | Not in our dataset | Not in our dataset |
| LiveCodeBenchCompetitive coding | 91.6% | Not in our dataset | Not in our dataset | Not in our dataset |
| Tau-BenchAgentic tool use | 89.1% | Not in our dataset | Not in our dataset | Not in our dataset |
| Terminal-BenchAgentic terminal | Not in our dataset | Not in our dataset | Not in our dataset | 75.1% |
| ARC-AGI-2Visual reasoning | 61.4% | 67.1% | Not in our dataset | 74.0% |
Pricing & context Published USD rates per 1M tokens
| Specification | V4 Flash Thinking | Grok 4.6 | Qwen3 Max Qwen 3.7 Max Preview | GPT-5.4 Thinking |
|---|---|---|---|---|
| Input | $0.091 | $2 | $0.78 | $2.5 |
| Output | $0.182 | $6 | $3.9 | $15 |
| Context window | 1.0M | 500K | 262K | 1.1M |
| 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.