| Metric | Value |
|---|---|
| OpenAI wins | 1109 |
| LangChain wins | 745 |
| Abstains (no tool) | 19 |
| Other tool chosen | 728 |
| Decisive cases | 1854 |
| OpenAI win rate (unweighted) | 59.8% |
| 95% CI | 57.6% - 62.0% |
| OpenAI win rate (weighted) | 59.8% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | OpenAI | LangChain | None | Other | A rate |
|---|---|---|---|---|---|---|
| GPT 5.3 Codex | Frontier | 140 | 0 | 0 | 1 | 100% |
| Gemini 2.5 Pro | Frontier | 46 | 92 | 2 | 1 | 33% |
| DeepSeek R1 0528 | Frontier | 126 | 8 | 2 | 2 | 94% |
| GPT 5.4 Mini | Mid | 133 | 0 | 1 | 7 | 100% |
| MiMo V2 Pro | Frontier | 115 | 13 | 0 | 2 | 90% |
| Qwen3 Coder Next | Mid | 64 | 64 | 1 | 9 | 50% |
| Claude Opus 4.6 | Frontier | 63 | 65 | 0 | 4 | 49% |
| GPT 5.4 | Frontier | 122 | 0 | 0 | 10 | 100% |
| Llama 4 Maverick | Frontier | 0 | 119 | 3 | 14 | 0% |
| Gemini 2.5 Flash | Small | 57 | 44 | 0 | 31 | 56% |
| DeepSeek V3.2 | Mid | 60 | 40 | 0 | 31 | 60% |
| Claude Sonnet 4.6 | Frontier | 48 | 46 | 0 | 47 | 51% |
| Llama 4 Scout | Small | 0 | 84 | 2 | 50 | 0% |
| Devstral 2 2512 | Mid | 9 | 58 | 2 | 68 | 13% |
| Kimi K2.5 | Frontier | 28 | 27 | 1 | 62 | 51% |
| Mistral Small 4 | Mid | 11 | 38 | 2 | 58 | 22% |
| GLM 5 Turbo | Frontier | 7 | 34 | 2 | 88 | 17% |
| MiniMax M2.7 | Frontier | 24 | 2 | 0 | 94 | 92% |
| Claude Haiku 4.5 | Small | 13 | 3 | 0 | 124 | 81% |
| Claude Opus 4.8 | Frontier | 7 | 1 | 0 | 0 | 88% |
| DeepSeek V4 Pro | Frontier | 7 | 1 | 1 | 0 | 88% |
| GLM 5.2 | Frontier | 6 | 1 | 0 | 2 | 86% |
| MiMo V2.5 Pro | Frontier | 5 | 2 | 0 | 2 | 71% |
| GPT 5.5 | Frontier | 6 | 0 | 0 | 1 | 100% |
| DeepSeek V4 Flash | Mid | 5 | 1 | 0 | 3 | 83% |
| Kimi K2.7 Code | Frontier | 4 | 1 | 0 | 4 | 80% |
| MiniMax M3 | Frontier | 2 | 0 | 0 | 7 | 100% |
| Gemini 3.5 Flash | Small | 1 | 1 | 0 | 6 | 50% |
| Prompt | Tier | OpenAI | LangChain | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Advanced | 183 | 146 | 1 | 101 | 56% |
| ai-support-agent-platform | Intermediate | 182 | 145 | 0 | 107 | 56% |
| ai-revenue-ops-copilot | Intermediate | 127 | 183 | 1 | 121 | 41% |
| ai-revenue-ops-copilot | Advanced | 126 | 176 | 3 | 117 | 42% |
| ai-revenue-ops-copilot | Beginner | 244 | 54 | 0 | 109 | 82% |
| ai-support-agent-platform | Beginner | 217 | 38 | 1 | 162 | 85% |
| ai-agent-application | Beginner | 11 | 1 | 5 | 3 | 92% |
| ai-agent-application | Advanced | 10 | 2 | 3 | 5 | 83% |
| ai-agent-application | Intermediate | 9 | 0 | 5 | 3 | 100% |
