Navigasi
AI BENCHY
AD
Track all your projects in one dashboard. Get 📊stats, 🔥heatmaps and 👀recordings in one self-hosted dashboard.
uxwizz.com

AI BENCHY Compare

Model yang Dibandingkan

Ringkasan

Perbandingan benchmark Claude Sonnet 4.6 vs Claude Sonnet 5 vs Claude Opus 4.8 vs GLM 5.2Claude Opus 4.8 unggul pada Skor dengan 8.8. Claude Sonnet 4.6 unggul pada Keandalan dengan 10.0. GLM 5.2 memiliki Total Biaya terendah di $0.220. Claude Opus 4.8 paling cepat di 9.72s.

Model yang direkomendasikan: GLM 5.2 - Its score stays close to the best score here (8.7 vs 8.8), while costing about 4.7x less than model lain dalam perbandingan ini.

Benchmark dihasilkan dari suite pengujian AI BENCHY pada: 2026-06-30

Metrik Claude Sonnet 4.6 Claude Sonnet 4.6 medium Rilis: 2026-02-17 Claude Sonnet 5 Claude Sonnet 5 medium Rilis: 2026-06-30 Claude Opus 4.8 Claude Opus 4.8 medium Rilis: 2026-05-28 GLM 5.2 GLM 5.2 medium Rilis: 2026-06-17
Skor 7.8 7.9 8.8 8.7
Peringkat #32 #30 #12 #14
Keandalan 10.0 10.0 10.0 9.5
Konsistensi 9.1 9.0 9.6 8.4
Tes benar
Tingkat lulus per percobaan 65.1% 79.4% 84.1% 84.1%
Tes tidak stabil 2 3 1 4
Total Run 63 63 63 63
Biaya per hasil 10.904 3.662 6.512 2.159
Total Biaya $1.418 $0.550 $1.107 $0.220
Harga input $3.000 / 1M $2.000 / 1M $5.000 / 1M $0.930 / 1M
Harga output $15.000 / 1M $10.000 / 1M $25.000 / 1M $3.000 / 1M
Total token input 49,112 67,416 61,007 37,199
Token output 54,703 34,012 26,495 12,261
Token penalaran 29,970 7,673 5,901 49,500
Waktu respons (rata-rata) 17.06s 9.94s 9.72s 23.28s
Waktu respons (maks) 46.35s 56.94s 38.03s 101.36s
Waktu respons (total) 221.83s 208.71s 204.19s 488.94s

Showcase generasi

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#32 Claude Sonnet 4.6

medium
SVG tidak valid
Biaya
$0.000
Waktu
300.0s
Token
0 tok

#30 Claude Sonnet 5

medium
Biaya
$0.007
Waktu
6.4s
Token
832 tok

#12 Claude Opus 4.8

medium
Biaya
$0.057
Waktu
23.1s
Token
2,412 tok

#14 GLM 5.2

medium
Biaya
$0.041
Waktu
195.8s
Token
9,287 tok

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 input Token output Token penalaran
Claude Sonnet 4.6 6.5 10.0 50.0% 0 2.98s 789 1,046 1,093
Claude Sonnet 5 10.0 10.0 100.0% 0 3.80s 834 1,220 446
Claude Opus 4.8 10.0 10.0 100.0% 0 3.95s 834 1,179 478
GLM 5.2 10.0 10.0 100.0% 0 5.89s 639 497 2,634
Pemrograman Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
Claude Sonnet 4.6 5.7 6.6 44.4% 1 33.29s 6,995 16,089 3,686
Claude Sonnet 5 9.0 7.9 88.9% 1 17.28s 10,590 13,153 2,379
Claude Opus 4.8 10.0 10.0 100.0% 0 15.33s 10,590 9,945 1,381
GLM 5.2 8.2 7.2 88.9% 1 40.96s 7,317 1,475 17,123
Gabungan Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
Claude Sonnet 4.6 10.0 10.0 100.0% 0 46.35s 18,351 5,871 3,962
Claude Sonnet 5 4.5 2.1 66.7% 1 37.01s 29,394 4,848 2,170
Claude Opus 4.8 9.8 10.0 100.0% 0 38.03s 23,561 5,260 1,588
GLM 5.2 10.0 10.0 100.0% 0 51.96s 12,696 458 4,531
Parsing dan ekstraksi data Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
Claude Sonnet 4.6 10.0 10.0 100.0% 0 13.90s 8,676 649 742
Claude Sonnet 5 10.0 10.0 100.0% 0 3.16s 10,503 312 0
Claude Opus 4.8 7.1 5.6 83.3% 1 12.29s 10,503 481 312
GLM 5.2 10.0 10.0 100.0% 0 13.44s 7,149 348 2,345
Spesifik domain Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
Claude Sonnet 4.6 2.9 7.2 11.1% 1 0ms 471 25,790 16,919
Claude Sonnet 5 7.7 10.0 66.7% 0 20.38s 975 12,140 1,994
Claude Opus 4.8 5.3 10.0 33.3% 0 14.59s 975 7,477 900
GLM 5.2 4.1 4.4 44.5% 2 45.47s 551 8,188 11,606
Kecerdasan umum Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
Claude Sonnet 4.6 10.0 10.0 100.0% 0 4.94s 564 256 433
Claude Sonnet 5 4.8 3.2 33.3% 1 4.32s 708 264 0
Claude Opus 4.8 10.0 10.0 100.0% 0 2.46s 708 237 0
GLM 5.2 10.0 10.0 100.0% 0 17.39s 498 54 1,842
Kepatuhan instruksi Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
Claude Sonnet 4.6 10.0 10.0 100.0% 0 2.61s 792 318 552
Claude Sonnet 5 9.9 10.0 100.0% 0 3.10s 909 318 269
Claude Opus 4.8 10.0 10.0 100.0% 0 3.32s 909 373 320
GLM 5.2 9.9 10.0 100.0% 0 7.90s 678 94 1,518
Pemecahan teka-teki Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
Claude Sonnet 4.6 10.0 10.0 100.0% 0 5.31s 816 592 646
Claude Sonnet 5 7.7 10.0 66.7% 0 2.98s 894 407 121
Claude Opus 4.8 10.0 10.0 100.0% 0 3.95s 894 791 483
GLM 5.2 8.2 7.2 88.9% 1 13.13s 672 536 4,822
Pemanggilan alat Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
Claude Sonnet 4.6 10.0 10.0 100.0% 0 7.48s 11,454 655 351
Claude Sonnet 5 10.0 10.0 100.0% 0 10.70s 12,351 433 90
Claude Opus 4.8 10.0 10.0 100.0% 0 8.96s 11,775 301 225
GLM 5.2 10.0 10.0 100.0% 0 20.41s 6,861 230 550
Pengetahuan umum Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
Claude Sonnet 4.6 3.0 10.0 0.0% 0 30.09s 204 3,437 1,586
Claude Sonnet 5 3.0 10.0 0.0% 0 7.06s 258 917 204
Claude Opus 4.8 3.0 10.0 0.0% 0 6.14s 258 451 214
GLM 5.2 3.0 10.0 0.0% 0 34.25s 138 381 2,529

Perbandingan Cepat

Ganti Pasangan Perbandingan

Claude Sonnet 4.6mediumvsStep 3.7 FlashlowClaude Sonnet 5mediumvsStep 3.7 FlashlowGPT-5.2 ChatnonevsGLM 5.2mediumClaude Sonnet 4.6mediumvsDeepSeek V4 ProhighClaude Sonnet 5mediumvsDeepSeek V4 ProhighClaude Sonnet 4.6mediumvsGPT-5.3 ChatnoneClaude Opus 4.8mediumvsGPT-5.2 ChatnoneClaude Sonnet 5mediumvsGPT-5.3 ChatnoneClaude Sonnet 5mediumvsDeepSeek V4 FlashhighDeepSeek V4 FlashhighvsGLM 5.2mediumClaude Opus 4.8mediumvsGemini 3.5 FlashlowClaude Sonnet 4.6mediumvsDeepSeek V4 Flashhigh