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 | Qwen | DeepSeek | Google | Anthropic |
|---|---|---|---|---|
| Arena ELOHuman preference | Not in our dataset | 1,436 | 1,433 | 1,414 |
| LiveBench 1.0General reasoning | Not in our dataset | 74.2% | Not in our dataset | 74.6% |
| GPQA DiamondScientific knowledge | 76.2% | 71.2% | 86.9% | 74.9% |
| HLEAcademic reasoning | Not in our dataset | 8.1% | 16.0% | 6.7% |
| MMLU ProMulti-task understanding | 79.1% | 83.0% | Not in our dataset | 86.6% |
| SimpleBenchCommonsense reasoning | Not in our dataset | 61.1% | Not in our dataset | 58.8% |
| AIME 2025Mathematics | Not in our dataset | Not in our dataset | Not in our dataset | 33.9% |
| AiderCode editing | Not in our dataset | Not in our dataset | Not in our dataset | 70.7% |
| SWE-Bench VerifiedAgentic coding | Not in our dataset | 73.7% | Not in our dataset | 72.5% |
| LiveCodeBenchCompetitive coding | Not in our dataset | 55.2% | 72.0% | 46.9% |
| Tau-BenchAgentic tool use | Not in our dataset | Not in our dataset | Not in our dataset | 81.8% |
| ARC-AGI-2Visual reasoning | Not in our dataset | 2.1% | Not in our dataset | 1.3% |
Pricing & context Published USD rates per 1M tokens
| Specification | 3.5 4B Thinking | V4 Flash | Gemini 3.1 Flash Lite Latest | Claude Opus 4 4.0 Non-thinking |
|---|---|---|---|---|
| Input | Not available | $0.091 | $0.25 | $15 |
| Output | Not available | $0.182 | $1.5 | $75 |
| Context window | Not available | 1.0M | Not available | 200K |
| 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.