Home / Compare
Compare 1-to-1
Put two models side by side on quality, price and privacy. For each criterion you can see which one wins.
Every pair, in one grid
Each cell shows the Elo difference of the row against the column, with a link to the comparison. Yellow: the row wins. "=" means within the measurement margin of 5 points.
 |  |  |  |  |  |  |  |  |  |  |  | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gemini | +21 | +40 | +42 | +42 | +43 | +44 | +52 | +56 | +101 | vs | vs | |
| Claude | −21 | +19 | +21 | +21 | +22 | +23 | +31 | +35 | +80 | vs | vs | |
| Kimi | −40 | −19 | = | = | = | = | +12 | +16 | +61 | vs | vs | |
| ChatGPT | −42 | −21 | = | = | = | = | +10 | +14 | +59 | vs | vs | |
| Copilot | −42 | −21 | = | = | = | = | +10 | +14 | +59 | vs | vs | |
| Qwen | −43 | −22 | = | = | = | = | +9 | +13 | +58 | vs | vs | |
| GLM (Z.ai) | −44 | −23 | = | = | = | = | +8 | +12 | +57 | vs | vs | |
| DeepSeek | −52 | −31 | −12 | −10 | −10 | −9 | −8 | = | +49 | vs | vs | |
| Grok | −56 | −35 | −16 | −14 | −14 | −13 | −12 | = | +45 | vs | vs | |
| Le Chat (Mistral) | −101 | −80 | −61 | −59 | −59 | −58 | −57 | −49 | −45 | vs | vs | |
| Perplexity | vs | vs | vs | vs | vs | vs | vs | vs | vs | vs | vs | |
| Llama | vs | vs | vs | vs | vs | vs | vs | vs | vs | vs | vs |