AI BENCHY Fouten
Instructies niet gevolgd-fouten
Zie welke AI-modellen het vaakst tegen Instructies niet gevolgd aanlopen, zodat je betrouwbaarheidsrisico's ziet voordat je kiest.
| Rang | Model | Bedrijf | Instructies niet gevolgd-aantal | Score | Correcte tests | Responstijd (gem.) |
|---|---|---|---|---|---|---|
| #38 | GPT-5.4 Nano medium | OpenAI | 3 | 7.6 | 11/18 | 11.2s |
| #40 | GPT-5.2 medium | OpenAI | 3 | 7.5 | 11/18 | 14.0s |
| #57 | GPT-5 Nano medium | OpenAI | 3 | 6.3 | 7/18 | 44.1s |
| #60 | Gemma 4 26B A4B none | 3 | 6.2 | 7/18 | 6.59s | |
| #69 | Kimi K2.6 none | Moonshot AI | 3 | 5.8 | 7/18 | 2.05s |
| #71 | MiniMax M2.5 medium | Minimax | 3 | 5.7 | 5/18 | 39.6s |
| #73 | Mistral Small 4 medium | Mistral | 3 | 5.7 | 5/18 | 5.64s |
| #79 | Grok 4.20 Beta none | X AI | 3 | 5.3 | 4/18 | 1.19s |
| #81 | Elephant medium | Openrouter | 3 | 5.2 | 5/18 | 1.27s |
| #85 | Elephant none | Openrouter | 3 | 5.2 | 5/18 | 1.23s |
| #86 | GPT-5.4 Mini none | OpenAI | 3 | 5.1 | 5/18 | 1.17s |
| #90 | Qwen3.5-9B none | Qwen | 3 | 4.8 | 4/18 | 1.47s |
| #96 | GPT-5.4 Nano none | OpenAI | 3 | 4.5 | 2/18 | 1.40s |
| #6 | Seed-2.0-Lite medium | Bytedance Seed | 2 | 8.6 | 13/18 | 30.4s |
| #7 | GPT-5.3-Codex medium | OpenAI | 2 | 8.6 | 13/18 | 15.4s |