| Metric | Value |
|---|---|
| OpenAI wins | 1699 |
| LangChain wins | 954 |
| Abstains (no tool) | 118 |
| Other tool chosen | 1034 |
| Decisive cases | 2653 |
| OpenAI win rate (unweighted) | 64.0% |
| 95% CI | 62.2% - 65.8% |
| OpenAI win rate (weighted) | 64.0% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model range | Tier | OpenAI | LangChain | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek Flash 3.2–4.1 | Mid | 101 | 46 | 9 | 42 | 69% |
| DeepSeek Pro 4–4 0813 | Frontier | 56 | 3 | 9 | 4 | 95% |
| DeepSeek R1 | Frontier | 167 | 11 | 21 | 2 | 94% |
| Devstral | Mid | 15 | 67 | 14 | 79 | 18% |
| Gemini Flash 2.5–3.8 | Small | 81 | 56 | 1 | 64 | 59% |
| Gemini Pro 2.5–3.1 Preview | Frontier | 82 | 116 | 2 | 3 | 41% |
| GLM 5 Turbo–5.3 | Frontier | 54 | 39 | 3 | 106 | 58% |
| GPT 5.4–6 Astra | Frontier | 179 | 0 | 0 | 19 | 100% |
| GPT Coding 5.3 Codex–6.1 Sol | Frontier | 196 | 0 | 0 | 7 | 100% |
| GPT Mini 5.4 Mini–6 Luna | Mid | 184 | 0 | 1 | 17 | 100% |
| Haiku | Small | 18 | 4 | 0 | 181 | 82% |
| Kimi 2.5–3 | Frontier | 58 | 28 | 3 | 99 | 67% |
| Llama Maverick | Frontier | 0 | 156 | 24 | 18 | 0% |
| Llama Scout | Small | 0 | 122 | 10 | 62 | 0% |
| MiMo Pro 2–2.6 | Frontier | 153 | 25 | 4 | 19 | 86% |
| MiniMax M 2.7–3 | Frontier | 58 | 5 | 0 | 128 | 92% |
| Mistral Small | Mid | 12 | 58 | 16 | 80 | 17% |
| Opus 4.6–5.5 | Frontier | 123 | 66 | 0 | 14 | 65% |
| Qwen Coder | Mid | 88 | 96 | 1 | 16 | 48% |
| Sonnet 4.6–5.5 | Frontier | 74 | 56 | 0 | 74 | 57% |
| Prompt | Tier | OpenAI | LangChain | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Advanced | 248 | 183 | 1 | 134 | 58% |
| ai-support-agent-platform | Intermediate | 245 | 185 | 0 | 140 | 57% |
| ai-revenue-ops-copilot | Beginner | 322 | 75 | 1 | 140 | 81% |
| ai-revenue-ops-copilot | Intermediate | 185 | 211 | 1 | 171 | 47% |
| ai-revenue-ops-copilot | Advanced | 174 | 218 | 4 | 156 | 44% |
| ai-support-agent-platform | Beginner | 295 | 53 | 1 | 198 | 85% |
| ai-agent-application | Advanced | 80 | 10 | 25 | 40 | 89% |
| ai-agent-application | Beginner | 78 | 12 | 45 | 23 | 87% |
| ai-agent-application | Intermediate | 72 | 7 | 40 | 32 | 91% |
