Coverage
Every demo against every model, under every harness. The gaps are the point: this is the queue for the next arcade run.
generatedprompt changed sinceproduced nothingfailednot generated
The cube
Three axes, because the harness is a variable too — a model that failed a demo under minimal-v1 may well clear it under a retuned harness, and both results are worth keeping. Drag to orbit, scroll to zoom, click a block to open it.
The matrix
The same thing flat, one table per harness. A cell is a link: filled ones open the result, empty ones open the demo.
minimal-v1 37/104 generated
| qwen3.6-27b | qwen3.6-27b-q6 | gemma-4-31b | gemma-4-12b | gemma-4-26b-a4b | qwen3.6-35b-a3b | bielik-11b-v3 | qwen3.5-122b-a10b | muse-glimmer-30b | nemotron-3.5-lightning-30b-a3b | qwen3.8-27b | qwen3.8-flash-next | qwen3.8-flash-next-strata | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Bounce — the full game desktop | |||||||||||||
| Bounce — Level 1 desktop | |||||||||||||
| Breakout desktop | |||||||||||||
| Fix the platformer desktop | |||||||||||||
| Snake, with one idea of your own desktop | |||||||||||||
| Analytics dashboard desktop | |||||||||||||
| Fix the WszystkieKina city schedule desktop | |||||||||||||
| Product landing page responsive |
claude-code-2.1.234 0/104 generated
| qwen3.6-27b | qwen3.6-27b-q6 | gemma-4-31b | gemma-4-12b | gemma-4-26b-a4b | qwen3.6-35b-a3b | bielik-11b-v3 | qwen3.5-122b-a10b | muse-glimmer-30b | nemotron-3.5-lightning-30b-a3b | qwen3.8-27b | qwen3.8-flash-next | qwen3.8-flash-next-strata | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Bounce — the full game desktop | |||||||||||||
| Bounce — Level 1 desktop | |||||||||||||
| Breakout desktop | |||||||||||||
| Fix the platformer desktop | |||||||||||||
| Snake, with one idea of your own desktop | |||||||||||||
| Analytics dashboard desktop | |||||||||||||
| Fix the WszystkieKina city schedule desktop | |||||||||||||
| Product landing page responsive |
opencode-1.18.18 0/104 generated
| qwen3.6-27b | qwen3.6-27b-q6 | gemma-4-31b | gemma-4-12b | gemma-4-26b-a4b | qwen3.6-35b-a3b | bielik-11b-v3 | qwen3.5-122b-a10b | muse-glimmer-30b | nemotron-3.5-lightning-30b-a3b | qwen3.8-27b | qwen3.8-flash-next | qwen3.8-flash-next-strata | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Bounce — the full game desktop | |||||||||||||
| Bounce — Level 1 desktop | |||||||||||||
| Breakout desktop | |||||||||||||
| Fix the platformer desktop | |||||||||||||
| Snake, with one idea of your own desktop | |||||||||||||
| Analytics dashboard desktop | |||||||||||||
| Fix the WszystkieKina city schedule desktop | |||||||||||||
| Product landing page responsive |
qwen-code-0.21.13 0/104 generated
| qwen3.6-27b | qwen3.6-27b-q6 | gemma-4-31b | gemma-4-12b | gemma-4-26b-a4b | qwen3.6-35b-a3b | bielik-11b-v3 | qwen3.5-122b-a10b | muse-glimmer-30b | nemotron-3.5-lightning-30b-a3b | qwen3.8-27b | qwen3.8-flash-next | qwen3.8-flash-next-strata | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Bounce — the full game desktop | |||||||||||||
| Bounce — Level 1 desktop | |||||||||||||
| Breakout desktop | |||||||||||||
| Fix the platformer desktop | |||||||||||||
| Snake, with one idea of your own desktop | |||||||||||||
| Analytics dashboard desktop | |||||||||||||
| Fix the WszystkieKina city schedule desktop | |||||||||||||
| Product landing page responsive |