| Metric | Value |
|---|---|
| Langfuse wins | 82 |
| Promptfoo wins | 98 |
| Abstains (no tool) | 194 |
| Other tool chosen | 3563 |
| Decisive cases | 180 |
| Langfuse win rate (unweighted) | 45.6% |
| 95% CI | 38.4% - 52.8% |
| Langfuse win rate (weighted) | 45.6% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Langfuse | Promptfoo | None | Other | A rate |
|---|---|---|---|---|---|---|
| Qwen3 Coder Next | Mid | 23 | 0 | 4 | 176 | 100% |
| GPT 5.4 Mini | Mid | 9 | 9 | 8 | 181 | 50% |
| Claude Sonnet 4.6 | Frontier | 16 | 0 | 7 | 187 | 100% |
| GLM 5.2 | Frontier | 0 | 16 | 0 | 68 | 0% |
| Claude Opus 4.8 | Frontier | 13 | 1 | 7 | 62 | 93% |
| Gemini 3.5 Flash | Small | 0 | 13 | 8 | 63 | 0% |
| Kimi K2.7 Code | Frontier | 0 | 13 | 1 | 69 | 0% |
| MiMo V2.5 Pro | Frontier | 2 | 9 | 9 | 63 | 18% |
| Mistral Small 4 | Mid | 1 | 10 | 3 | 182 | 9% |
| DeepSeek V4 Flash | Mid | 4 | 5 | 4 | 70 | 44% |
| MiMo V2 Pro | Frontier | 0 | 6 | 8 | 112 | 0% |
| Claude Haiku 4.5 | Small | 5 | 0 | 4 | 191 | 100% |
| GLM 5 Turbo | Frontier | 0 | 5 | 19 | 102 | 0% |
| Gemini 2.5 Pro | Frontier | 4 | 0 | 9 | 195 | 100% |
| Llama 4 Scout | Small | 4 | 0 | 23 | 162 | 100% |
| DeepSeek V4 Pro | Frontier | 0 | 4 | 2 | 78 | 0% |
| MiniMax M3 | Frontier | 0 | 3 | 6 | 75 | 0% |
| Kimi K2.5 | Frontier | 0 | 2 | 3 | 109 | 0% |
| DeepSeek V3.2 | Mid | 1 | 0 | 22 | 100 | 100% |
| GPT 5.3 Codex | Frontier | 0 | 1 | 0 | 209 | 0% |
| GPT 5.5 | Frontier | 0 | 1 | 3 | 80 | 0% |
| Claude Opus 4.6 | Frontier | 0 | 0 | 0 | 126 | n/a |
| DeepSeek R1 0528 | Frontier | 0 | 0 | 10 | 200 | n/a |
| Devstral 2 2512 | Mid | 0 | 0 | 14 | 158 | n/a |
| Gemini 2.5 Flash | Small | 0 | 0 | 1 | 120 | n/a |
| GPT 5.4 | Frontier | 0 | 0 | 0 | 126 | n/a |
| Llama 4 Maverick | Frontier | 0 | 0 | 14 | 181 | n/a |
| MiniMax M2.7 | Frontier | 0 | 0 | 5 | 118 | n/a |
| Prompt | Tier | Langfuse | Promptfoo | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Beginner | 29 | 5 | 75 | 419 | 85% |
| ai-engineering-workflow | Beginner | 5 | 20 | 61 | 48 | 20% |
| ai-engineering-workflow | Intermediate | 9 | 14 | 1 | 107 | 39% |
| ai-revenue-ops-copilot | Intermediate | 9 | 11 | 4 | 499 | 45% |
| ai-engineering-workflow | Advanced | 3 | 17 | 1 | 109 | 15% |
| ai-support-agent-platform | Intermediate | 10 | 7 | 5 | 500 | 59% |
| ai-support-agent-platform | Advanced | 7 | 6 | 5 | 509 | 54% |
| ai-revenue-ops-copilot | Beginner | 5 | 5 | 11 | 504 | 50% |
| ai-revenue-ops-copilot | Advanced | 0 | 7 | 2 | 507 | 0% |
| ai-agent-application | Beginner | 4 | 2 | 28 | 104 | 67% |
| ai-agent-application | Intermediate | 1 | 3 | 1 | 127 | 25% |
| ai-agent-application | Advanced | 0 | 1 | 0 | 130 | 0% |
