Navigasi
AI BENCHY
Your ad here

AI BENCHY Compare

Model yang Dibandingkan

Benchmark dihasilkan dari suite pengujian AI BENCHY pada: 2026-04-14

Metrik Elephant Elephant none Rilis: 2026-04-14 Qwen3.5-27B Qwen3.5-27B none Rilis: 2026-02-24 Elephant Elephant medium Rilis: 2026-04-14 Qwen3.5-27B Qwen3.5-27B medium Rilis: 2026-02-24
Skor 5.2 5.9 5.2 8.4
Peringkat #81 #64 #77 #8
Konsistensi 9.6 9.2 9.6 8.8
Tes benar
Tingkat lulus per percobaan 31.5% 38.9% 29.6% 81.5%
Tes tidak stabil 1 2 1 3
Total Run 54 54 54 54
Biaya per hasil 0.000 0.265 0.000 3.822
Total Biaya $0.000 $0.016 $0.000 $0.497
Harga input $0.000 / 1M $0.195 / 1M $0.000 / 1M $0.195 / 1M
Harga output $0.000 / 1M $1.560 / 1M $0.000 / 1M $1.560 / 1M
Token output 2,573 3,545 2,596 2,500
Token penalaran 0 0 0 242,500
Waktu respons (rata-rata) 1.23s 1.74s 1.27s 53.03s
Waktu respons (maks) 3.81s 9.39s 3.70s 163.96s
Waktu respons (total) 22.16s 31.32s 22.82s 954.46s

Model teratas berdasarkan skor

Skor vs Total Biaya

Waktu respons (rata-rata)

Skor vs Waktu respons (rata-rata)

Total token output

Skor vs Total token output

Rincian Kategori

Trik anti-AI Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token output Token penalaran
Elephant 6.6 10.0 50.0% 0 963ms 610 0
Qwen3.5-27B 4.8 10.0 25.0% 0 788ms 267 0
Elephant 6.6 10.0 50.0% 0 1.19s 815 0
Qwen3.5-27B 8.7 7.9 91.7% 1 19.75s 569 31,505
Pemrograman Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token output Token penalaran
Elephant 6.4 3.3 66.7% 1 1.39s 375 0
Qwen3.5-27B 10.0 10.0 100.0% 0 2.51s 381 0
Elephant 5.1 3.3 33.3% 1 1.30s 365 0
Qwen3.5-27B 10.0 10.0 100.0% 0 70.35s 375 19,165
Gabungan Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token output Token penalaran
Elephant 3.0 10.0 0.0% 0 3.81s 731 0
Qwen3.5-27B 2.8 1.6 33.3% 1 9.39s 1,461 0
Elephant 3.0 10.0 0.0% 0 3.70s 562 0
Qwen3.5-27B 10.0 10.0 100.0% 0 163.96s 483 9,991
Parsing dan ekstraksi data Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token output Token penalaran
Elephant 6.5 10.0 50.0% 0 1.04s 246 0
Qwen3.5-27B 10.0 10.0 100.0% 0 1.43s 243 0
Elephant 6.5 10.0 50.0% 0 979ms 246 0
Qwen3.5-27B 10.0 10.0 100.0% 0 30.26s 270 16,150
Spesifik domain Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token output Token penalaran
Elephant 3.0 10.0 0.0% 0 927ms 24 0
Qwen3.5-27B 3.0 10.0 0.0% 0 540ms 15 0
Elephant 3.0 10.0 0.0% 0 925ms 24 0
Qwen3.5-27B 5.3 10.0 33.3% 0 79.53s 43 52,368
Kecerdasan umum Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token output Token penalaran
Elephant 4.0 10.0 0.0% 0 854ms 106 0
Qwen3.5-27B 5.0 10.0 0.0% 0 2.51s 126 0
Elephant 4.3 10.0 0.0% 0 920ms 105 0
Qwen3.5-27B 6.1 3.1 66.7% 1 101.41s 70 23,147
Kepatuhan instruksi Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token output Token penalaran
Elephant 9.8 10.0 100.0% 0 1.03s 81 0
Qwen3.5-27B 4.8 10.0 0.0% 0 815ms 69 0
Elephant 9.8 10.0 100.0% 0 987ms 82 0
Qwen3.5-27B 10.0 10.0 100.0% 0 19.66s 97 11,638
Pemecahan teka-teki Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token output Token penalaran
Elephant 3.3 10.0 0.0% 0 849ms 170 0
Qwen3.5-27B 6.7 7.9 55.6% 1 1.37s 680 0
Elephant 3.7 10.0 0.0% 0 867ms 166 0
Qwen3.5-27B 8.2 7.7 77.8% 1 64.61s 245 77,213
Pemanggilan alat Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token output Token penalaran
Elephant 3.0 10.0 0.0% 0 2.79s 230 0
Qwen3.5-27B 10.0 10.0 100.0% 0 3.54s 303 0
Elephant 3.0 10.0 0.0% 0 2.83s 231 0
Qwen3.5-27B 10.0 10.0 100.0% 0 7.45s 348 1,323

Perbandingan Cepat

Ganti Pasangan Perbandingan

ElephantmediumvsGrok 4.20nonegpt-oss-120bmediumTersedia gratisvsQwen3.5-27BnoneMistral Small 4nonevsElephantmediumgpt-oss-120bnoneTersedia gratisvsElephantmediumMiniMax M2.7mediumvsElephantnoneTrinity Large PreviewnoneTersedia gratisvsElephantmediumGPT-5.4 MininonevsElephantmediumElephantmediumvsQwen3 Coder NextnoneNemotron 3 SupernoneTersedia gratisvsElephantmediumMiniMax M2.5mediumTersedia gratisvsQwen3.5-27BnoneMistral Small 4mediumvsQwen3.5-27BnoneElephantmediumvsGLM 5 Turbonone