Leading: OpenRouter (90.7%)
| Metric | Value |
|---|---|
| Helicone wins | 18 |
| OpenRouter wins | 176 |
| Abstains (no tool) | 8 |
| Other tool chosen | 788 |
| Decisive cases | 194 |
| Helicone win rate (unweighted) | 9.3% |
| 95% CI | 5.9% - 14.2% |
| Helicone win rate (weighted) | 9.3% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model range | Tier | Helicone | OpenRouter | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek Flash 3.2–4.1 | Mid | 0 | 6 | 0 | 48 | 0% |
| DeepSeek Pro 4–4 0813 | Frontier | 0 | 4 | 0 | 50 | 0% |
| DeepSeek R1 | Frontier | 2 | 4 | 1 | 47 | 33% |
| Devstral | Mid | 0 | 5 | 0 | 35 | 0% |
| Gemini Flash 2.5–3.8 | Small | 0 | 8 | 0 | 46 | 0% |
| Gemini Pro 2.5–3.1 Preview | Frontier | 0 | 0 | 0 | 54 | n/a |
| GLM 5 Turbo–5.3 | Frontier | 0 | 16 | 0 | 38 | 0% |
| GPT 5.4–6 Astra | Frontier | 0 | 13 | 0 | 37 | 0% |
| GPT Coding 5.3 Codex–6.1 Sol | Frontier | 0 | 16 | 0 | 37 | 0% |
| GPT Mini 5.4 Mini–6 Luna | Mid | 0 | 18 | 1 | 35 | 0% |
| Haiku | Small | 16 | 4 | 0 | 21 | 80% |
| Kimi 2.5–3 | Frontier | 0 | 5 | 0 | 37 | 0% |
| Llama Maverick | Frontier | 0 | 0 | 1 | 53 | n/a |
| Llama Scout | Small | 0 | 0 | 0 | 30 | n/a |
| MiMo Pro 2–2.6 | Frontier | 0 | 11 | 0 | 42 | 0% |
| MiniMax M 2.7–3 | Frontier | 0 | 8 | 0 | 43 | 0% |
| Mistral Small | Mid | 0 | 0 | 5 | 38 | n/a |
| Opus 4.6–5.5 | Frontier | 0 | 21 | 0 | 33 | 0% |
| Qwen Coder | Mid | 0 | 27 | 0 | 20 | 0% |
| Sonnet 4.6–5.5 | Frontier | 0 | 10 | 0 | 44 | 0% |
| Prompt | Tier | Helicone | OpenRouter | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-engineering-workflow | Beginner | 0 | 100 | 6 | 66 | 0% |
| ai-agent-application | Beginner | 0 | 36 | 1 | 134 | 0% |
| ai-agent-application | Intermediate | 1 | 32 | 1 | 133 | 3% |
| ai-agent-application | Advanced | 7 | 4 | 0 | 154 | 64% |
| ai-engineering-workflow | Advanced | 9 | 1 | 0 | 146 | 90% |
| ai-engineering-workflow | Intermediate | 1 | 3 | 0 | 155 | 25% |
