Leading: LangChain (79.3%)
| Metric | Value |
|---|---|
| LangGraph wins | 74 |
| LangChain wins | 283 |
| Abstains (no tool) | 4 |
| Other tool chosen | 636 |
| Decisive cases | 357 |
| LangGraph win rate (unweighted) | 20.7% |
| 95% CI | 16.8% - 25.2% |
| LangGraph win rate (weighted) | 20.7% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | LangGraph | LangChain | None | Other | A rate |
|---|---|---|---|---|---|---|
| Llama 4 Maverick | Frontier | 0 | 45 | 0 | 8 | 0% |
| GLM 5 Turbo | Frontier | 22 | 16 | 2 | 14 | 58% |
| Gemini 2.5 Pro | Frontier | 0 | 37 | 1 | 16 | 0% |
| Kimi K2.5 | Frontier | 24 | 8 | 1 | 14 | 75% |
| Claude Sonnet 4.6 | Frontier | 13 | 19 | 0 | 22 | 41% |
| Claude Opus 4.6 | Frontier | 0 | 27 | 0 | 27 | 0% |
| Llama 4 Scout | Small | 0 | 27 | 0 | 26 | 0% |
| Mistral Small 4 | Mid | 12 | 12 | 0 | 18 | 50% |
| Qwen3 Coder Next | Mid | 2 | 21 | 0 | 31 | 9% |
| Devstral 2 2512 | Mid | 0 | 22 | 0 | 30 | 0% |
| Gemini 2.5 Flash | Small | 0 | 19 | 0 | 35 | 0% |
| DeepSeek V3.2 | Mid | 0 | 15 | 0 | 38 | 0% |
| MiMo V2 Pro | Frontier | 0 | 9 | 0 | 44 | 0% |
| DeepSeek R1 0528 | Frontier | 0 | 4 | 0 | 49 | 0% |
| Claude Haiku 4.5 | Small | 0 | 2 | 0 | 52 | 0% |
| MiniMax M2.7 | Frontier | 1 | 0 | 0 | 51 | 100% |
| GPT 5.3 Codex | Frontier | 0 | 0 | 0 | 54 | n/a |
| GPT 5.4 | Frontier | 0 | 0 | 0 | 54 | n/a |
| GPT 5.4 Mini | Mid | 0 | 0 | 0 | 53 | n/a |
| Prompt | Tier | LangGraph | LangChain | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-revenue-ops-copilot | Intermediate | 30 | 73 | 0 | 64 | 29% |
| ai-revenue-ops-copilot | Advanced | 13 | 71 | 2 | 78 | 15% |
| ai-support-agent-platform | Intermediate | 14 | 58 | 0 | 99 | 19% |
| ai-support-agent-platform | Advanced | 15 | 55 | 1 | 99 | 21% |
| ai-revenue-ops-copilot | Beginner | 1 | 14 | 0 | 145 | 7% |
| ai-support-agent-platform | Beginner | 1 | 12 | 1 | 151 | 8% |