Leading: Playwright (68.5%)
| Metric | Value |
|---|---|
| Browserbase wins | 85 |
| Playwright wins | 185 |
| Abstains (no tool) | 112 |
| Other tool chosen | 5 |
| Decisive cases | 270 |
| Browserbase win rate (unweighted) | 31.5% |
| 95% CI | 26.2% - 37.2% |
| Browserbase win rate (weighted) | 31.5% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Browserbase | Playwright | None | Other | A rate |
|---|---|---|---|---|---|---|
| GPT 5.4 Mini | Mid | 2 | 18 | 1 | 0 | 10% |
| DeepSeek V4 Flash | Mid | 0 | 17 | 4 | 0 | 0% |
| Mistral Small 4 | Mid | 0 | 17 | 4 | 0 | 0% |
| MiniMax M3 | Frontier | 14 | 2 | 2 | 2 | 88% |
| GPT 5.3 Codex | Frontier | 1 | 15 | 5 | 0 | 6% |
| Gemini 3.5 Flash | Small | 14 | 1 | 5 | 1 | 93% |
| GPT 5.5 | Frontier | 13 | 2 | 6 | 0 | 87% |
| Kimi K2.7 Code | Frontier | 10 | 5 | 6 | 0 | 67% |
| DeepSeek V4 Pro | Frontier | 4 | 11 | 6 | 0 | 27% |
| Claude Haiku 4.5 | Small | 5 | 9 | 7 | 0 | 36% |
| MiMo V2.5 Pro | Frontier | 5 | 9 | 6 | 1 | 36% |
| Claude Opus 4.8 | Frontier | 4 | 10 | 7 | 0 | 29% |
| Claude Sonnet 4.6 | Frontier | 0 | 14 | 7 | 0 | 0% |
| DeepSeek R1 0528 | Frontier | 0 | 14 | 7 | 0 | 0% |
| Llama 4 Maverick | Frontier | 0 | 14 | 7 | 0 | 0% |
| GLM 5.2 | Frontier | 12 | 1 | 7 | 1 | 92% |
| Devstral 2 2512 | Mid | 0 | 9 | 5 | 0 | 0% |
| Llama 4 Scout | Small | 0 | 7 | 7 | 0 | 0% |
| Qwen3 Coder Next | Mid | 0 | 6 | 7 | 0 | 0% |
| Gemini 2.5 Pro | Frontier | 1 | 4 | 6 | 0 | 20% |
| Prompt | Tier | Browserbase | Playwright | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-agent-application | Intermediate | 42 | 83 | 1 | 2 | 34% |
| ai-agent-application | Advanced | 37 | 84 | 0 | 1 | 31% |
| ai-agent-application | Beginner | 6 | 18 | 111 | 2 | 25% |
