Leading: Intercom (63.1%)
| Metric | Value |
|---|---|
| Anthropic Claude API wins | 76 |
| Intercom wins | 130 |
| Abstains (no tool) | 128 |
| Other tool chosen | 3654 |
| Decisive cases | 206 |
| Anthropic Claude API win rate (unweighted) | 36.9% |
| 95% CI | 30.6% - 43.7% |
| Anthropic Claude API win rate (weighted) | 36.9% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model range | Tier | Anthropic Claude API | Intercom | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek Flash 3.2–4.1 | Mid | 0 | 16 | 9 | 183 | 0% |
| DeepSeek Pro 4–4 0813 | Frontier | 0 | 0 | 9 | 72 | n/a |
| DeepSeek R1 | Frontier | 0 | 1 | 23 | 186 | 0% |
| Devstral | Mid | 0 | 2 | 16 | 167 | 0% |
| Gemini Flash 2.5–3.8 | Small | 0 | 0 | 1 | 210 | n/a |
| Gemini Pro 2.5–3.1 Preview | Frontier | 0 | 2 | 2 | 208 | 0% |
| GLM 5 Turbo–5.3 | Frontier | 0 | 2 | 3 | 206 | 0% |
| GPT 5.4–6 Astra | Frontier | 0 | 9 | 0 | 198 | 0% |
| GPT Coding 5.3 Codex–6.1 Sol | Frontier | 0 | 0 | 0 | 212 | n/a |
| GPT Mini 5.4 Mini–6 Luna | Mid | 0 | 9 | 1 | 201 | 0% |
| Haiku | Small | 49 | 9 | 0 | 154 | 84% |
| Kimi 2.5–3 | Frontier | 0 | 1 | 3 | 193 | 0% |
| Llama Maverick | Frontier | 0 | 6 | 27 | 173 | 0% |
| Llama Scout | Small | 0 | 23 | 11 | 169 | 0% |
| MiMo Pro 2–2.6 | Frontier | 0 | 4 | 4 | 202 | 0% |
| MiniMax M 2.7–3 | Frontier | 16 | 8 | 0 | 179 | 67% |
| Mistral Small | Mid | 0 | 21 | 18 | 135 | 0% |
| Opus 4.6–5.5 | Frontier | 0 | 0 | 0 | 212 | n/a |
| Qwen Coder | Mid | 0 | 0 | 1 | 209 | n/a |
| Sonnet 4.6–5.5 | Frontier | 11 | 17 | 0 | 185 | 39% |
| Prompt | Tier | Anthropic Claude API | Intercom | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Beginner | 3 | 129 | 1 | 434 | 2% |
| ai-support-agent-platform | Advanced | 38 | 0 | 1 | 550 | 100% |
| ai-revenue-ops-copilot | Advanced | 22 | 0 | 4 | 547 | 100% |
| ai-support-agent-platform | Intermediate | 6 | 1 | 0 | 584 | 86% |
| ai-revenue-ops-copilot | Intermediate | 5 | 0 | 1 | 582 | 100% |
| ai-revenue-ops-copilot | Beginner | 2 | 0 | 1 | 553 | 100% |
| ai-agent-application | Advanced | 0 | 0 | 26 | 149 | n/a |
| ai-agent-application | Beginner | 0 | 0 | 50 | 128 | n/a |
| ai-agent-application | Intermediate | 0 | 0 | 44 | 127 | n/a |