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 | Qwen | Cohere | Meta | Microsoft |
|---|---|---|---|---|
| Arena ELOHuman preference | Not in our dataset | 1,227 | 1,211 | Not in our dataset |
| LiveBench 1.0General reasoning | 21.8% | Not in our dataset | 26.0% | 25.3% |
| GPQA DiamondScientific knowledge | 22.9% | Not in our dataset | 32.8% | 25.2% |
| MMLU ProMulti-task understanding | Not in our dataset | Not in our dataset | Not in our dataset | 52.8% |
| SimpleBenchCommonsense reasoning | Not in our dataset | 17.4% | Not in our dataset | Not in our dataset |
| AIME 2025Mathematics | 2.6% | Not in our dataset | 2.7% | 5.3% |
| LiveCodeBenchCompetitive coding | 3.6% | Not in our dataset | 10.8% | 10.4% |
Pricing & context Published USD rates per 1M tokens
| Specification | Qwen3-0.6B (Thinking) Non-thinking | Command R Non-thinking | Llama 3 8B 3.1 (Non-thinking) | Phi-4 Mini Non-Thinking |
|---|---|---|---|---|
| Input | Not available | $0.5 | $0.05 | $0.08 |
| Output | Not available | $1.5 | $0.08 | $0.35 |
| Context window | Not available | 128K | 131K | 128K |
| 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.