Kegagalan AI BENCHY
Kegagalan Tidak mengikuti instruksi
Lihat model AI mana yang paling sering mengalami Tidak mengikuti instruksi, agar Anda bisa melihat risiko keandalan sebelum memilih. Urutkan berdasarkan: Waktu respons (rata-rata) ↑.
| Peringkat | Model | Perusahaan | Jumlah Tidak mengikuti instruksi | Skor | Tes benar | Waktu respons (rata-rata) |
|---|---|---|---|---|---|---|
| #142 | Mistral Small 4 none | Mistral | 1 | 4.9 | 5/21 | 630ms |
| #155 | Mercury 2 none | Inception | 1 | 4.5 | 4/21 | 653ms |
| #163 | Granite 4.1 8B none | IBM Granite | 4 | 4.0 | 2/21 | 728ms |
| #162 | Nemotron 3 Nano Omni 30b A3b Reasoning none | NVIDIA | 2 | 4.1 | 2/19 | 728ms |
| #160 | LFM2-24B-A2B none | Liquid | 1 | 4.2 | 2/16 | 782ms |
| #146 | Laguna Xs.2 none | Poolside | 1 | 4.8 | 5/19 | 806ms |
| #90 | Gemini 3.1 Flash Lite none | 1 | 6.4 | 9/21 | 1.06s | |
| #144 | GPT-5.4 Mini none | OpenAI | 3 | 4.9 | 5/21 | 1.13s |
| #106 | Grok 4.20 Beta none | X AI | 1 | 5.8 | 6/18 | 1.19s |
| #58 | Gemini 3.1 Flash Lite Preview none | 2 | 7.2 | 12/21 | 1.21s | |
| #137 | Elephant Alpha none | Openrouter | 3 | 5.1 | 5/21 | 1.22s |
| #136 | Elephant Alpha medium | Openrouter | 2 | 5.1 | 6/21 | 1.27s |
| #87 | Gemini 3.1 Flash Lite minimal | 3 | 6.4 | 10/21 | 1.33s | |
| #125 | GPT-5.4 none | OpenAI | 1 | 5.5 | 7/21 | 1.42s |
| #148 | GPT-5.4 Nano none | OpenAI | 2 | 4.7 | 4/21 | 1.48s |