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 DeepSeek V4 Flash vs DeepSeek V4 Pro vs Kimi K2.6 vs GLM 5GLM 5 unggul pada Skor dengan 8.6. DeepSeek V4 Flash unggul pada Keandalan dengan 10.0. DeepSeek V4 Flash memiliki Total Biaya terendah di $0.027. GLM 5 paling cepat di 33.54s.

Model yang direkomendasikan: DeepSeek V4 Flash - Its score stays close to the best score here (8.3 vs 8.6), while costing about 16.2x less than model lain dalam perbandingan ini.

Benchmark dihasilkan dari suite pengujian AI BENCHY pada: 2026-06-18

Metrik DeepSeek V4 Flash DeepSeek V4 Flash high Rilis: 2026-04-24 DeepSeek V4 Pro DeepSeek V4 Pro high Rilis: 2026-04-24 Kimi K2.6 Kimi K2.6 medium Rilis: 2026-04-20 Tersedia gratis GLM 5 GLM 5 medium Rilis: 2026-02-12
Skor 8.3 7.6 7.8 8.6
Peringkat #23 #41 #35 #15
Keandalan 10.0 9.3 10.0 10.0
Konsistensi 8.5 7.0 8.6 8.5
Tes benar
Tingkat lulus per percobaan 74.6% 66.7% 65.1% 82.5%
Tes tidak stabil 4 8 3 4
Total Run 63 63 63 63
Biaya per hasil 0.299 1.742 8.358 1.668
Total Biaya $0.027 $0.157 $0.889 $0.228
Harga input $0.090 / 1M $0.435 / 1M $0.680 / 1M $0.600 / 1M
Harga output $0.180 / 1M $0.870 / 1M $3.410 / 1M $1.920 / 1M
Total token input 39,745 38,726 29,450 35,224
Token output 10,310 6,334 102,923 21,570
Token penalaran 123,501 159,151 254,094 102,996
Waktu respons (rata-rata) 45.85s 77.20s 71.67s 33.54s
Waktu respons (maks) 218.13s 416.76s 406.78s 99.85s
Waktu respons (total) 962.79s 1621.17s 1433.36s 435.99s

Showcase generasi

Hamster playing table tennis

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

#23 DeepSeek V4 Flash

high
Biaya
$0.003
Waktu
93.1s
Token
7,926 tok

#41 DeepSeek V4 Pro

high
Biaya
$0.023
Waktu
257.6s
Token
14,870 tok

#35 MoonshotAI: Kimi K2.6

medium
Biaya
$0.013
Waktu
103.4s
Token
3,620 tok

#15 GLM 5

medium
Biaya
$0.005
Waktu
20.7s
Token
2,068 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
DeepSeek V4 Flash 8.3 10.0 75.0% 0 28.51s 540 140 7,770
DeepSeek V4 Pro 5.7 5.9 58.3% 2 25.70s 536 149 3,214
Kimi K2.6 7.0 8.0 66.7% 1 11.59s 618 7,115 8,934
GLM 5 10.0 10.0 100.0% 0 23.66s 555 480 7,056
Pemrograman Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
DeepSeek V4 Flash 7.8 10.0 66.7% 0 50.60s 7,279 395 34,862
DeepSeek V4 Pro 6.1 4.6 66.7% 2 243.00s 5,090 383 84,580
Kimi K2.6 5.7 8.6 33.3% 0 214.42s 2,925 9,970 77,189
GLM 5 10.0 10.0 100.0% 0 74.30s 7,254 2,997 52,930
Gabungan Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
DeepSeek V4 Flash 10.0 10.0 100.0% 0 76.57s 14,016 465 7,347
DeepSeek V4 Pro 10.0 10.0 100.0% 0 38.17s 14,060 454 5,836
Kimi K2.6 10.0 10.0 100.0% 0 40.96s 11,271 711 13,876
GLM 5 10.0 10.0 100.0% 0 28.96s 12,804 662 3,242
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
DeepSeek V4 Flash 10.0 10.0 100.0% 0 28.03s 7,290 201 1,179
DeepSeek V4 Pro 10.0 10.0 100.0% 0 25.03s 7,690 274 2,166
Kimi K2.6 10.0 10.0 100.0% 0 20.38s 7,014 316 11,305
GLM 5 7.1 5.6 83.3% 1 8.90s 5,508 567 3,734
Spesifik domain Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
DeepSeek V4 Flash 4.1 4.4 44.5% 2 100.31s 666 27 59,249
DeepSeek V4 Pro 3.6 7.2 22.2% 1 151.46s 569 4,404 50,391
Kimi K2.6 5.3 7.2 44.4% 1 202.38s 326 47,035 98,262
GLM 5 3.5 4.4 33.3% 2 0ms 260 13,176 14,137
Kecerdasan umum Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
DeepSeek V4 Flash 6.1 3.1 66.7% 1 25.15s 471 79 632
DeepSeek V4 Pro 10.0 10.0 100.0% 0 8.83s 471 115 1,013
Kimi K2.6 10.0 10.0 100.0% 0 17.83s 477 3,981 4,472
GLM 5 6.1 3.1 66.7% 1 14.69s 477 2,020 2,248
Kepatuhan instruksi Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
DeepSeek V4 Flash 10.0 10.0 100.0% 0 15.36s 627 63 1,622
DeepSeek V4 Pro 7.8 6.6 83.3% 1 8.73s 627 66 2,726
Kimi K2.6 10.0 10.0 100.0% 0 12.53s 669 3,977 5,269
GLM 5 10.0 10.0 100.0% 0 7.25s 636 1,001 2,129
Pemecahan teka-teki Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
DeepSeek V4 Flash 8.2 7.2 88.9% 1 26.11s 594 196 1,767
DeepSeek V4 Pro 6.9 4.9 77.8% 2 56.85s 591 178 2,563
Kimi K2.6 6.0 7.4 55.6% 1 25.06s 651 13,860 17,599
GLM 5 10.0 10.0 100.0% 0 11.33s 609 33 4,076
Pemanggilan alat Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
DeepSeek V4 Flash 10.0 10.0 100.0% 0 74.73s 8,079 228 542
DeepSeek V4 Pro 9.8 10.0 100.0% 0 15.92s 8,909 295 701
Kimi K2.6 10.0 10.0 100.0% 0 8.92s 5,286 248 1,011
GLM 5 10.0 10.0 100.0% 0 15.93s 6,935 233 994
Pengetahuan umum Skor Konsistensi Tingkat lulus per percobaan Tes tidak stabil Tes benar Waktu respons (rata-rata) Token input Token output Token penalaran
DeepSeek V4 Flash 3.0 10.0 0.0% 0 54.46s 183 8,516 8,531
DeepSeek V4 Pro 3.0 10.0 0.0% 0 34.01s 183 16 5,961
Kimi K2.6 3.0 10.0 0.0% 0 130.27s 213 15,710 16,177
GLM 5 3.0 10.0 0.0% 0 67.37s 186 401 12,450

Perbandingan Cepat

Ganti Pasangan Perbandingan

DeepSeek V4 ProhighvsMiniMax M3mediumDeepSeek V4 ProhighvsGrok Build 0.1mediumDeepSeek V4 FlashhighvsGemini 2.5 FlashmediumKimi K2.6mediumTersedia gratisvsStep 3.7 FlashlowDeepSeek V4 ProhighvsKimi K2.5mediumDeepSeek V4 ProhighvsMercury 2mediumGPT-5.2 ChatnonevsGLM 5mediumDeepSeek V4 ProhighvsStep 3.7 FlashlowDeepSeek V4 FlashhighvsQwen3.7 PlusmediumClaude Opus 4.6mediumvsDeepSeek V4 ProhighDeepSeek V4 ProhighvsGPT-5.3 ChatnoneDeepSeek V4 ProhighvsGPT-5.4 Nanomedium