| Metric | Value |
|---|---|
| LangChain wins | 709 |
| OpenAI wins | 1027 |
| Abstains (no tool) | 6 |
| Other tool chosen | 686 |
| Decisive cases | 1736 |
| LangChain win rate (unweighted) | 40.8% |
| 95% CI | 38.6% - 43.2% |
| LangChain win rate (weighted) | 40.8% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | LangChain | OpenAI | None | Other | A rate |
|---|---|---|---|---|---|---|
| GPT 5.3 Codex | Frontier | 0 | 132 | 0 | 0 | 0% |
| Gemini 2.5 Pro | Frontier | 87 | 42 | 2 | 1 | 67% |
| Claude Opus 4.6 | Frontier | 65 | 63 | 0 | 4 | 51% |
| MiMo V2 Pro | Frontier | 13 | 115 | 0 | 2 | 10% |
| DeepSeek R1 0528 | Frontier | 8 | 119 | 0 | 2 | 6% |
| GPT 5.4 Mini | Mid | 0 | 124 | 1 | 7 | 0% |
| GPT 5.4 | Frontier | 0 | 122 | 0 | 10 | 0% |
| Qwen3 Coder Next | Mid | 59 | 61 | 0 | 9 | 49% |
| Llama 4 Maverick | Frontier | 114 | 0 | 0 | 13 | 100% |
| Gemini 2.5 Flash | Small | 44 | 57 | 0 | 31 | 44% |
| DeepSeek V3.2 | Mid | 40 | 60 | 0 | 31 | 40% |
| Claude Sonnet 4.6 | Frontier | 43 | 43 | 0 | 46 | 50% |
| Llama 4 Scout | Small | 80 | 0 | 0 | 47 | 100% |
| Devstral 2 2512 | Mid | 54 | 9 | 0 | 66 | 86% |
| Kimi K2.5 | Frontier | 27 | 28 | 1 | 62 | 49% |
| Mistral Small 4 | Mid | 36 | 10 | 0 | 56 | 78% |
| GLM 5 Turbo | Frontier | 34 | 7 | 2 | 88 | 83% |
| MiniMax M2.7 | Frontier | 2 | 24 | 0 | 94 | 8% |
| Claude Haiku 4.5 | Small | 3 | 11 | 0 | 117 | 21% |
| Prompt | Tier | LangChain | OpenAI | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Advanced | 139 | 175 | 1 | 96 | 44% |
| ai-support-agent-platform | Intermediate | 137 | 175 | 0 | 102 | 44% |
| ai-revenue-ops-copilot | Intermediate | 178 | 120 | 1 | 113 | 60% |
| ai-revenue-ops-copilot | Advanced | 169 | 119 | 3 | 112 | 59% |
| ai-revenue-ops-copilot | Beginner | 51 | 232 | 0 | 105 | 18% |
| ai-support-agent-platform | Beginner | 35 | 206 | 1 | 158 | 15% |
