Output Loop Bench · Marketing
Plumber flyer (A5)
Why we chose this prompt: A print piece a tradesperson would really hand out. It shows whether a model can build clear hierarchy, an offer and trust signals that read at arm's length.
| Model | Maker | Time | Tokens |
|---|---|---|---|
| claude-opus-5-5 ★ Output Loop's pick | Anthropic | 181.1s | 106,117 |
| muse-spark-1.3-medium | Meta | 16.1s | 24,842 |
| nemotron-3-nano-4b | NVIDIA | 16.5s | 1,984 |
| composer-2.5 | Cursor | 19.1s | 22,945 |
| gemini-3.8-flash | 32.8s | 21,627 | |
| claude-haiku-5-5 | Anthropic | 41.4s | 99,598 |
| gemma-4-26b-a4b-qat | 45.3s | 5,718 | |
| grok-4.7-build-fast | xAI | 49.6s | 23,967 |
| gemini-4-argon-medium | 53.0s | 22,201 | |
| kimi-k3-max | Moonshot (Kimi) | 54.3s | 29,047 |
| glm-5.3 | Z.ai | 59.3s | 19,269 |
| qwen3.8-flash | Alibaba (Qwen) | 59.7s | 21,568 |
| deepseek-v4-pro | DeepSeek | 65.1s | 22,927 |
| muse-glimmer-30b | Meta | 65.3s | 18,631 |
| MiniMax-M2.5 | MiniMax | 66.0s | 21,820 |
| gpt-6-astra | OpenAI | 75.3s | 17,625 |
| gpt-6.1-sol | OpenAI | 91.8s | 18,016 |
| kimi-k2.7-code | Moonshot (Kimi) | 92.0s | 39,194 |
| qwen3.7-plus | Alibaba (Qwen) | 96.1s | 22,073 |
| claude-fable-5-1 | Anthropic | 102.8s | 97,797 |
| claude-sonnet-5-5 | Anthropic | 117.1s | 105,726 |
| muse-glimmer | Meta | 122.4s | 4,154 |
| bonsai-27b | Prism ML | 129.3s | 7,305 |
| gemini-3.1-pro | 130.2s | 22,642 | |
| deepseek-v4.1-flash | DeepSeek | 170.9s | 36,559 |
| kimi-k3 | Moonshot (Kimi) | 282.5s | 17,849 |
| gpt-6-luna | OpenAI | 293.4s | 30,815 |
| grok-4.7 | xAI | 340.7s | 43,942 |
| qwen3.8-max | Alibaba (Qwen) | 527.6s | 40,012 |
| nemotron-3.5-lightning-30b-a3b | NVIDIA | 552.5s | 222,755 |
| qwen3.8-27b | Alibaba (Qwen) | 980.9s | 30,786 |