Leading: LangGraph (65.6%)
| Metric | Value |
|---|---|
| Intercom wins | 97 |
| LangGraph wins | 185 |
| Abstains (no tool) | 6 |
| Other tool chosen | 2140 |
| Decisive cases | 282 |
| Intercom win rate (unweighted) | 34.4% |
| 95% CI | 29.1% - 40.1% |
| Intercom win rate (weighted) | 34.4% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Intercom | LangGraph | None | Other | A rate |
|---|---|---|---|---|---|---|
| Kimi K2.5 | Frontier | 1 | 60 | 1 | 56 | 2% |
| GLM 5 Turbo | Frontier | 0 | 58 | 2 | 71 | 0% |
| Claude Sonnet 4.6 | Frontier | 13 | 33 | 0 | 86 | 28% |
| Mistral Small 4 | Mid | 14 | 26 | 0 | 62 | 35% |
| Llama 4 Scout | Small | 19 | 0 | 0 | 108 | 100% |
| DeepSeek V3.2 | Mid | 13 | 0 | 0 | 118 | 100% |
| GPT 5.4 | Frontier | 9 | 1 | 0 | 122 | 90% |
| MiniMax M2.7 | Frontier | 8 | 1 | 0 | 111 | 89% |
| GPT 5.4 Mini | Mid | 7 | 0 | 1 | 124 | 100% |
| Llama 4 Maverick | Frontier | 5 | 0 | 0 | 122 | 100% |
| Claude Haiku 4.5 | Small | 4 | 0 | 0 | 127 | 100% |
| Claude Opus 4.6 | Frontier | 0 | 4 | 0 | 128 | 0% |
| MiMo V2 Pro | Frontier | 2 | 0 | 0 | 128 | 100% |
| Qwen3 Coder Next | Mid | 0 | 2 | 0 | 127 | 0% |
| DeepSeek R1 0528 | Frontier | 1 | 0 | 0 | 128 | 100% |
| Gemini 2.5 Pro | Frontier | 1 | 0 | 2 | 129 | 100% |
| Devstral 2 2512 | Mid | 0 | 0 | 0 | 129 | n/a |
| Gemini 2.5 Flash | Small | 0 | 0 | 0 | 132 | n/a |
| GPT 5.3 Codex | Frontier | 0 | 0 | 0 | 132 | n/a |
| Prompt | Tier | Intercom | LangGraph | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Beginner | 97 | 5 | 1 | 297 | 95% |
| ai-revenue-ops-copilot | Intermediate | 0 | 70 | 1 | 341 | 0% |
| ai-revenue-ops-copilot | Advanced | 0 | 44 | 3 | 356 | 0% |
| ai-support-agent-platform | Advanced | 0 | 32 | 1 | 378 | 0% |
| ai-support-agent-platform | Intermediate | 0 | 30 | 0 | 384 | 0% |
| ai-revenue-ops-copilot | Beginner | 0 | 4 | 0 | 384 | 0% |
