Peringkat Gabungan
Lihat model AI mana yang paling baik di Gabungan, mana yang tetap andal, dan di mana kesenjangan terbesar muncul.
Alasan kegagalan
Dengan alasan kegagalan Pemanggilan alat tidak valid180 Dengan alasan kegagalan Jawaban salah128 Dengan alasan kegagalan Kesalahan API62 Dengan alasan kegagalan Tidak ada jawaban56 Dengan alasan kegagalan Kedaluwarsa11 Dengan alasan kegagalan Format tambahan1 Dengan alasan kegagalan Tidak mengikuti instruksi1
451/451
Filter model
Tidak ada model yang cocok dengan pencarian dan filter saat ini.
| Peringkat | Model | Perusahaan | Skor Gabungan | Skor | Total Biaya | Tes benar | Waktu respons (rata-rata) |
|---|---|---|---|---|---|---|---|
| #153 | DeepSeek V4 Flash 0731 low | DeepSeek | 1.0 | 0.7 | $0.306 | 2/2 | 54.2s |
| #86 | Gemini 3.1 Pro Preview medium | 1.0 | 0.8 | $1.585 | 2/2 | 40.4s | |
| #147 | GLM 5.1 medium | Z.ai | 1.0 | 0.8 | $1.546 | 2/2 | 175.9s |
| #200 | Claude Opus 4.8 none | Anthropic | 1.0 | 0.7 | $2.334 | 2/2 | 26.4s |
| #283 | Claude Sonnet 4.6 none | Anthropic | 1.0 | 0.6 | $0.661 | 2/2 | 37.5s |
| #35 | Qwen3.8 MAX medium | Qwen | 0.9 | 0.9 | $1.181 | 1/2 | 91.3s |
| #37 | Qwen3.8 Max (0902) high | Qwen | 0.9 | 0.9 | $3.306 | 1/2 | 110.4s |
| #73 | Claude Sonnet 5.5 xhigh | Anthropic | 0.9 | 0.9 | $1.516 | 1/2 | 30.4s |
| #80 | Qwen3.7 Max medium | Qwen | 0.9 | 0.9 | $1.689 | 1/2 | 287.8s |
| #91 | GPT-5.6 Terra high | OpenAI | 0.9 | 0.8 | $0.698 | 1/2 | 13.7s |
| #113 | Gemini 3.8 Flash medium | 0.9 | 0.8 | $0.769 | 1/2 | 60.3s | |
| #115 | GPT-5.6 Terra low | OpenAI | 0.9 | 0.8 | $0.435 | 1/2 | 9.68s |
| #170 | Qwen3.8 27B medium | Qwen | 0.9 | 0.7 | ~$0.073 | 1/2 | 124.5s |
| #183 | MiMo-V2.5 medium | Xiaomi | 0.9 | 0.7 | $0.130 | 1/2 | 78.0s |
| #224 | Step 3.7 Flash high | Stepfun | 0.9 | 0.7 | $1.478 | 1/2 | 41.2s |