Leading: Anthropic (64.9%)
| Metric | Value |
|---|---|
| Anthropic Claude API wins | 52 |
| Anthropic wins | 96 |
| Abstains (no tool) | 19 |
| Other tool chosen | 2434 |
| Decisive cases | 148 |
| Anthropic Claude API win rate (unweighted) | 35.1% |
| 95% CI | 27.9% - 43.1% |
| Anthropic Claude API win rate (weighted) | 35.1% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Anthropic Claude API | Anthropic | None | Other | A rate |
|---|---|---|---|---|---|---|
| Claude Haiku 4.5 | Small | 36 | 60 | 0 | 44 | 38% |
| MiniMax M2.7 | Frontier | 16 | 18 | 0 | 86 | 47% |
| Mistral Small 4 | Mid | 0 | 10 | 2 | 97 | 0% |
| Kimi K2.7 Code | Frontier | 0 | 2 | 0 | 7 | 0% |
| DeepSeek V4 Flash | Mid | 0 | 1 | 0 | 8 | 0% |
| Gemini 3.5 Flash | Small | 0 | 1 | 0 | 7 | 0% |
| GLM 5 Turbo | Frontier | 0 | 1 | 2 | 128 | 0% |
| GPT 5.5 | Frontier | 0 | 1 | 0 | 6 | 0% |
| Kimi K2.5 | Frontier | 0 | 1 | 1 | 116 | 0% |
| MiniMax M3 | Frontier | 0 | 1 | 0 | 8 | 0% |
| Claude Opus 4.6 | Frontier | 0 | 0 | 0 | 132 | n/a |
| Claude Opus 4.8 | Frontier | 0 | 0 | 0 | 8 | n/a |
| Claude Sonnet 4.6 | Frontier | 0 | 0 | 0 | 141 | n/a |
| DeepSeek R1 0528 | Frontier | 0 | 0 | 2 | 136 | n/a |
| DeepSeek V3.2 | Mid | 0 | 0 | 0 | 131 | n/a |
| DeepSeek V4 Pro | Frontier | 0 | 0 | 1 | 8 | n/a |
| Devstral 2 2512 | Mid | 0 | 0 | 2 | 135 | n/a |
| Gemini 2.5 Flash | Small | 0 | 0 | 0 | 132 | n/a |
| Gemini 2.5 Pro | Frontier | 0 | 0 | 2 | 139 | n/a |
| GLM 5.2 | Frontier | 0 | 0 | 0 | 9 | n/a |
| GPT 5.3 Codex | Frontier | 0 | 0 | 0 | 141 | n/a |
| GPT 5.4 | Frontier | 0 | 0 | 0 | 132 | n/a |
| GPT 5.4 Mini | Mid | 0 | 0 | 1 | 140 | n/a |
| Llama 4 Maverick | Frontier | 0 | 0 | 3 | 133 | n/a |
| Llama 4 Scout | Small | 0 | 0 | 2 | 134 | n/a |
| MiMo V2 Pro | Frontier | 0 | 0 | 0 | 130 | n/a |
| MiMo V2.5 Pro | Frontier | 0 | 0 | 0 | 9 | n/a |
| Qwen3 Coder Next | Mid | 0 | 0 | 1 | 137 | n/a |
| Prompt | Tier | Anthropic Claude API | Anthropic | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Advanced | 28 | 9 | 1 | 393 | 76% |
| ai-support-agent-platform | Intermediate | 4 | 24 | 0 | 406 | 14% |
| ai-revenue-ops-copilot | Intermediate | 2 | 25 | 1 | 404 | 7% |
| ai-revenue-ops-copilot | Advanced | 16 | 4 | 3 | 399 | 80% |
| ai-revenue-ops-copilot | Beginner | 1 | 18 | 0 | 388 | 5% |
| ai-support-agent-platform | Beginner | 1 | 10 | 1 | 406 | 9% |
| ai-agent-application | Intermediate | 0 | 2 | 5 | 10 | 0% |
| ai-agent-application | Advanced | 0 | 2 | 3 | 15 | 0% |
| ai-agent-application | Beginner | 0 | 2 | 5 | 13 | 0% |