Leading: TruLens (60.0%)
| Metric | Value |
|---|---|
| Patronus AI wins | 2 |
| TruLens wins | 3 |
| Abstains (no tool) | 36 |
| Other tool chosen | 959 |
| Decisive cases | 5 |
| Patronus AI win rate (unweighted) | 40.0% |
| 95% CI | 11.8% - 76.9% |
| Patronus AI win rate (weighted) | 40.0% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Patronus AI | TruLens | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek R1 0528 | Frontier | 0 | 2 | 2 | 50 | 0% |
| Gemini 2.5 Pro | Frontier | 1 | 0 | 6 | 47 | 100% |
| MiMo V2 Pro | Frontier | 1 | 0 | 2 | 51 | 100% |
| Mistral Small 4 | Mid | 0 | 1 | 0 | 50 | 0% |
| Claude Haiku 4.5 | Small | 0 | 0 | 1 | 51 | n/a |
| Claude Opus 4.6 | Frontier | 0 | 0 | 0 | 54 | n/a |
| Claude Sonnet 4.6 | Frontier | 0 | 0 | 0 | 54 | n/a |
| DeepSeek V3.2 | Mid | 0 | 0 | 9 | 43 | n/a |
| Devstral 2 2512 | Mid | 0 | 0 | 1 | 50 | n/a |
| Gemini 2.5 Flash | Small | 0 | 0 | 0 | 52 | n/a |
| GLM 5 Turbo | Frontier | 0 | 0 | 7 | 47 | n/a |
| 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 | 1 | 53 | n/a |
| Kimi K2.5 | Frontier | 0 | 0 | 0 | 48 | n/a |
| Llama 4 Maverick | Frontier | 0 | 0 | 0 | 53 | n/a |
| Llama 4 Scout | Small | 0 | 0 | 3 | 46 | n/a |
| MiniMax M2.7 | Frontier | 0 | 0 | 1 | 51 | n/a |
| Qwen3 Coder Next | Mid | 0 | 0 | 3 | 51 | n/a |
| Prompt | Tier | Patronus AI | TruLens | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Beginner | 1 | 0 | 25 | 143 | 100% |
| ai-support-agent-platform | Advanced | 1 | 0 | 1 | 167 | 100% |
| ai-revenue-ops-copilot | Intermediate | 0 | 1 | 1 | 162 | 0% |
| ai-revenue-ops-copilot | Advanced | 0 | 1 | 1 | 162 | 0% |
| ai-support-agent-platform | Intermediate | 0 | 1 | 4 | 161 | 0% |
| ai-revenue-ops-copilot | Beginner | 0 | 0 | 4 | 164 | n/a |
