Leading: LangChain (55.8%)
| Metric | Value |
|---|---|
| GitHub Copilot wins | 247 |
| LangChain wins | 312 |
| Abstains (no tool) | 715 |
| Other tool chosen | 1171 |
| Decisive cases | 559 |
| GitHub Copilot win rate (unweighted) | 44.2% |
| 95% CI | 40.1% - 48.3% |
| GitHub Copilot win rate (weighted) | 44.2% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | GitHub Copilot | LangChain | None | Other | A rate |
|---|---|---|---|---|---|---|
| Gemini 2.5 Pro | Frontier | 86 | 0 | 18 | 28 | 100% |
| Llama 4 Scout | Small | 64 | 9 | 41 | 15 | 88% |
| Gemini 2.5 Flash | Small | 22 | 49 | 59 | 2 | 31% |
| MiMo V2 Pro | Frontier | 14 | 57 | 30 | 31 | 20% |
| DeepSeek R1 0528 | Frontier | 7 | 47 | 51 | 27 | 13% |
| Llama 4 Maverick | Frontier | 10 | 39 | 52 | 31 | 20% |
| Claude Opus 4.6 | Frontier | 0 | 38 | 20 | 74 | 0% |
| Qwen3 Coder Next | Mid | 0 | 33 | 17 | 79 | 0% |
| MiniMax M2.7 | Frontier | 15 | 10 | 25 | 79 | 60% |
| GPT 5.3 Codex | Frontier | 17 | 0 | 25 | 90 | 100% |
| DeepSeek V3.2 | Mid | 0 | 17 | 52 | 61 | 0% |
| Devstral 2 2512 | Mid | 5 | 9 | 55 | 56 | 36% |
| Kimi K2.5 | Frontier | 5 | 0 | 53 | 60 | 100% |
| Mistral Small 4 | Mid | 1 | 3 | 54 | 60 | 25% |
| GPT 5.4 Mini | Mid | 1 | 1 | 7 | 122 | 50% |
| Claude Haiku 4.5 | Small | 0 | 0 | 65 | 55 | n/a |
| Claude Sonnet 4.6 | Frontier | 0 | 0 | 58 | 74 | n/a |
| GLM 5 Turbo | Frontier | 0 | 0 | 2 | 130 | n/a |
| GPT 5.4 | Frontier | 0 | 0 | 31 | 97 | n/a |
| Prompt | Tier | GitHub Copilot | LangChain | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-revenue-ops-copilot | Intermediate | 96 | 38 | 2 | 264 | 72% |
| ai-support-agent-platform | Intermediate | 39 | 78 | 205 | 92 | 33% |
| ai-revenue-ops-copilot | Beginner | 32 | 53 | 222 | 108 | 38% |
| ai-revenue-ops-copilot | Advanced | 20 | 65 | 7 | 297 | 24% |
| ai-support-agent-platform | Advanced | 36 | 40 | 0 | 334 | 47% |
| ai-support-agent-platform | Beginner | 24 | 38 | 279 | 76 | 39% |
