Benchmark Results
Comparing different LLMs vibe coding the Przeprogramowani.pl website
Generated: 2/8/2026, 8:55:57 PM • Total attempts: 15
Model Family Rankings
1
GPT-5.3-Codex
3 attempts
Average
9.5/10.0
95.0%
2
Claude Opus 4.6
3 attempts
Average
8.5/10.0
85.0%
3
Gemini 3 Pro
3 attempts
Average
6.2/10.0
61.7%
4
Kimi K2.5
3 attempts
Average
5.7/10.0
56.7%
5
GLM-4.7
3 attempts
Average
0.2/10.0
1.7%
Detailed Comparison
Click on any score to reveal the detailed scoring explanation for that criterion.
| Criterion | GPT-5.3-Codex Attempt 1 | GPT-5.3-Codex Attempt 2 | GPT-5.3-Codex Attempt 3 | Claude Opus 4.6 Attempt 1 | Claude Opus 4.6 Attempt 2 | Claude Opus 4.6 Attempt 3 | Gemini 3 Pro Attempt 1 | Gemini 3 Pro Attempt 2 | Gemini 3 Pro Attempt 3 | Kimi K2.5 Attempt 1 | Kimi K2.5 Attempt 2 | Kimi K2.5 Attempt 3 | GLM-4.7 Attempt 1 | GLM-4.7 Attempt 2 | GLM-4.7 Attempt 3 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Task completion time | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A |
Local build | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 0 | 0 | 0 |
Manual testing | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
Tech stack | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0.5 | 0.5 | 0.5 | 0 | 0 | 0 | 0.5 |
O nas page | 1 | 1 | 0.5 | 1 | 0.5 | 1 | 1 | 1 | 0.5 | 0.5 | 1 | 0 | 0 | 0 | 0 |
Podcast page | 1 | 1 | 1 | 0.5 | 0.5 | 1 | 0.5 | 0.5 | 0.5 | 0 | 0 | 0 | 0 | 0 | 0 |
YouTube page | 1 | 1 | 1 | 0.5 | 0 | 1 | 0.5 | 0.5 | 0 | 0 | 0.5 | 0 | 0 | 0 | 0 |
Kursy section | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0.5 | 0 | 0 | 0 |
Consistent UI | 1 | 1 | 1 | 1 | 0.5 | 1 | 0.5 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
Responsive design | 0.5 | 1 | 0.5 | 1 | 0.5 | 0.5 | 1 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
SEO Tags | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0.5 | 1 | 1 | 0.5 | 0 | 0 | 0 |
Click on score cells to view evaluation notes
Score Legend
1.0 Full points - criterion met perfectly
0.5 Partial points - criterion mostly met
0.0 No points - criterion not met