Leading: Weights & Biases (56.3%)
| Metric | Value |
|---|---|
| Langfuse wins | 93 |
| Weights & Biases wins | 120 |
| Abstains (no tool) | 216 |
| Other tool chosen | 3852 |
| Decisive cases | 213 |
| Langfuse win rate (unweighted) | 43.7% |
| 95% CI | 37.2% - 50.4% |
| Langfuse win rate (weighted) | 43.7% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model range | Tier | Langfuse | Weights & Biases | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek Flash 3.2–4.1 | Mid | 5 | 0 | 26 | 192 | 100% |
| DeepSeek Pro 4–4 0813 | Frontier | 1 | 0 | 2 | 93 | 100% |
| DeepSeek R1 | Frontier | 0 | 10 | 11 | 207 | 0% |
| Devstral | Mid | 0 | 34 | 17 | 138 | 0% |
| Gemini Flash 2.5–3.8 | Small | 0 | 41 | 10 | 172 | 0% |
| Gemini Pro 2.5–3.1 Preview | Frontier | 4 | 1 | 14 | 207 | 80% |
| GLM 5 Turbo–5.3 | Frontier | 1 | 0 | 19 | 208 | 100% |
| GPT 5.4–6 Astra | Frontier | 1 | 0 | 4 | 223 | 100% |
| GPT Coding 5.3 Codex–6.1 Sol | Frontier | 0 | 0 | 1 | 227 | n/a |
| GPT Mini 5.4 Mini–6 Luna | Mid | 9 | 1 | 8 | 207 | 90% |
| Haiku | Small | 5 | 0 | 4 | 209 | 100% |
| Kimi 2.5–3 | Frontier | 0 | 0 | 5 | 209 | n/a |
| Llama Maverick | Frontier | 1 | 0 | 16 | 194 | 100% |
| Llama Scout | Small | 6 | 23 | 26 | 149 | 21% |
| MiMo Pro 2–2.6 | Frontier | 2 | 10 | 17 | 198 | 17% |
| MiniMax M 2.7–3 | Frontier | 0 | 0 | 12 | 213 | n/a |
| Mistral Small | Mid | 2 | 0 | 3 | 208 | 100% |
| Opus 4.6–5.5 | Frontier | 13 | 0 | 8 | 206 | 100% |
| Qwen Coder | Mid | 26 | 0 | 4 | 190 | 100% |
| Sonnet 4.6–5.5 | Frontier | 17 | 0 | 9 | 202 | 100% |
| Prompt | Tier | Langfuse | Weights & Biases | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Beginner | 33 | 12 | 77 | 443 | 73% |
| ai-revenue-ops-copilot | Advanced | 0 | 36 | 3 | 514 | 0% |
| ai-support-agent-platform | Advanced | 7 | 21 | 5 | 533 | 25% |
| ai-revenue-ops-copilot | Intermediate | 10 | 15 | 4 | 532 | 40% |
| ai-support-agent-platform | Intermediate | 10 | 14 | 6 | 531 | 42% |
| ai-revenue-ops-copilot | Beginner | 6 | 14 | 12 | 531 | 30% |
| ai-engineering-workflow | Intermediate | 10 | 3 | 1 | 136 | 77% |
| ai-engineering-workflow | Advanced | 5 | 5 | 1 | 138 | 50% |
| ai-engineering-workflow | Beginner | 6 | 0 | 72 | 75 | 100% |
| ai-agent-application | Beginner | 4 | 0 | 34 | 120 | 100% |
| ai-agent-application | Advanced | 1 | 0 | 0 | 150 | 100% |
| ai-agent-application | Intermediate | 1 | 0 | 1 | 149 | 100% |
