AI BENCHY

Leaderboard Benchmark AI

Pantau model AI SOTA teratas dengan leaderboard benchmark AI BENCHY—cara mudah untuk melihat model mana yang memimpin saat ini dalam skor, kualitas penalaran, keandalan, dan value. Urutkan berdasarkan: Tes benar ↓.

Benchmark dihasilkan dari suite pengujian AI BENCHY pada: 2026-03-06 Model Dievaluasi: 55

Cari 0/0

Peringkat	Model	Perusahaan	Skor	Biaya per hasil	Total Biaya	Waktu respons (rata-rata)
🥇 #1	Gemini 3 Flash Previewmedium	Google	10.0	1.025	$0.164	12.36s
Lihat kartu model Total Tes: 16 Tes Salah: 0 Tingkat lulus per percobaan: 100.0% Tes tidak stabil: 0 Token output: 1,634 Token penalaran: 47,907 Waktu respons: rata-rata 12.36s · total 111.21s · maks 50.16s Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 10.0 Tool Calling: 10.0
🥈 #2	Gemini 3.1 Pro Previewmedium	Google	9.4	3.417	$0.513	16.60s
Lihat kartu model Total Tes: 16 Tes Salah: 1 Tingkat lulus per percobaan: 93.8% Tes tidak stabil: 0 Token output: 1,521 Token penalaran: 35,656 Waktu respons: rata-rata 16.60s · total 149.36s · maks 40.61s Jawaban salah: 1 Anti-AI Tricks: 10.0 Combined: 9.0 Data parsing and extraction: 9.9 Domain specific: 7.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 10.0 Tool Calling: 10.0
🥉 #3	Qwen3.5 Plus 2026-02-15medium	Qwen	8.3	1.264	$0.165	34.45s
Lihat kartu model Total Tes: 16 Tes Salah: 3 Tingkat lulus per percobaan: 85.4% Tes tidak stabil: 1 Token output: 1,735 Token penalaran: 77,212 Waktu respons: rata-rata 34.45s · total 310.09s · maks 79.86s Kedaluwarsa: 2 Jawaban salah: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 10.0 Tool Calling: 10.0
#4	Gemini 3 Flash Previewlow	Google	8.2	0.607	$0.079	6.11s
Lihat kartu model Total Tes: 16 Tes Salah: 3 Tingkat lulus per percobaan: 83.3% Tes tidak stabil: 1 Token output: 1,586 Token penalaran: 19,950 Waktu respons: rata-rata 6.11s · total 97.74s · maks 14.72s Jawaban salah: 3 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 10.0 Instructions following: 9.5 Puzzle Solving: 10.0 Tool Calling: 10.0
#5	Gemini 3 Pro Previewmedium	Google	8.2	1.477	$0.192	7.15s
Lihat kartu model Total Tes: 16 Tes Salah: 3 Tingkat lulus per percobaan: 81.3% Tes tidak stabil: 0 Token output: 1,502 Token penalaran: 9,706 Waktu respons: rata-rata 7.15s · total 64.34s · maks 11.96s Jawaban salah: 3 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 10.0 Instructions following: 9.5 Puzzle Solving: 10.0 Tool Calling: 10.0
#6	GPT-5.3-Codexmedium	OpenAI	8.4	4.485	$0.539	16.59s
Lihat kartu model Total Tes: 16 Tes Salah: 4 Tingkat lulus per percobaan: 83.3% Tes tidak stabil: 2 Token output: 1,764 Token penalaran: 33,348 Waktu respons: rata-rata 16.59s · total 265.39s · maks 100.93s Tidak mengikuti instruksi: 2 Jawaban salah: 2 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 4.0 Instructions following: 10.0 Puzzle Solving: 9.3 Tool Calling: 10.0
#7	Qwen3.5-27Bmedium	Qwen	8.2	3.585	$0.431	52.13s
Lihat kartu model Total Tes: 16 Tes Salah: 4 Tingkat lulus per percobaan: 81.3% Tes tidak stabil: 2 Token output: 1,658 Token penalaran: 200,786 Waktu respons: rata-rata 52.13s · total 834.16s · maks 163.96s Tidak mengikuti instruksi: 2 Kedaluwarsa: 1 Jawaban salah: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 5.0 Instructions following: 10.0 Puzzle Solving: 8.3 Tool Calling: 10.0
#8	Gemini 3.1 Flash Lite Previewhigh	Google	8.2	19.243	$2.310	68.83s
Lihat kartu model Total Tes: 16 Tes Salah: 4 Tingkat lulus per percobaan: 77.1% Tes tidak stabil: 1 Token output: 1,283 Token penalaran: 1,533,310 Waktu respons: rata-rata 68.83s · total 1101.32s · maks 280.52s Jawaban salah: 3 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 10.0 Instructions following: 9.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#9	GPT-5.4medium	OpenAI	8.0	6.601	$0.793	20.05s
Lihat kartu model Total Tes: 16 Tes Salah: 4 Tingkat lulus per percobaan: 83.3% Tes tidak stabil: 3 Token output: 1,756 Token penalaran: 46,642 Waktu respons: rata-rata 20.05s · total 320.87s · maks 100.41s Tidak mengikuti instruksi: 2 Jawaban salah: 2 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 5.0 Instructions following: 10.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#10	Qwen3.5-122B-A10Bmedium	Qwen	7.7	4.095	$0.492	29.74s
Lihat kartu model Total Tes: 16 Tes Salah: 4 Tingkat lulus per percobaan: 79.2% Tes tidak stabil: 2 Token output: 17,292 Token penalaran: 145,625 Waktu respons: rata-rata 29.74s · total 475.83s · maks 119.29s Jawaban salah: 3 Kedaluwarsa: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 10.0 Tool Calling: 10.0
#11	Claude Sonnet 4.6medium	Anthropic	7.7	8.525	$1.023	11.23s
Lihat kartu model Total Tes: 16 Tes Salah: 4 Tingkat lulus per percobaan: 77.1% Tes tidak stabil: 1 Token output: 35,159 Token penalaran: 24,687 Waktu respons: rata-rata 11.23s · total 89.84s · maks 46.35s Format tambahan: 2 Kedaluwarsa: 1 Jawaban salah: 1 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 10.0 Tool Calling: 10.0
#12	Gemini 3.1 Flash Lite Previewmedium	Google	7.5	0.443	$0.049	3.83s
Lihat kartu model Total Tes: 16 Tes Salah: 5 Tingkat lulus per percobaan: 68.8% Tes tidak stabil: 0 Token output: 1,731 Token penalaran: 25,821 Waktu respons: rata-rata 3.83s · total 61.25s · maks 14.93s Jawaban salah: 4 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 9.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#13	GLM 5medium	Z.ai	7.4	0.933	$0.103	16.16s
Lihat kartu model Total Tes: 16 Tes Salah: 5 Tingkat lulus per percobaan: 83.3% Tes tidak stabil: 4 Token output: 19,773 Token penalaran: 36,459 Waktu respons: rata-rata 16.16s · total 129.26s · maks 28.96s Jawaban salah: 2 Tidak mengikuti instruksi: 1 Tidak ada jawaban: 1 Kedaluwarsa: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 5.0 Domain specific: 10.0 General Intelligence: 5.0 Instructions following: 10.0 Puzzle Solving: 10.0 Tool Calling: 10.0
#14	GPT-5.2 Chatnone	OpenAI	7.4	2.440	$0.269	7.03s
Lihat kartu model Total Tes: 16 Tes Salah: 5 Tingkat lulus per percobaan: 75.0% Tes tidak stabil: 2 Token output: 15,845 Token penalaran: 0 Waktu respons: rata-rata 7.03s · total 112.51s · maks 38.52s Jawaban salah: 4 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 4.0 Instructions following: 6.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#15	Gemini 2.5 Flashmedium	Google	7.4	2.619	$0.288	12.35s
Lihat kartu model Total Tes: 16 Tes Salah: 5 Tingkat lulus per percobaan: 72.9% Tes tidak stabil: 1 Token output: 1,370 Token penalaran: 110,522 Waktu respons: rata-rata 12.35s · total 197.62s · maks 95.48s Jawaban salah: 4 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 7.3 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 4.0 Instructions following: 9.5 Puzzle Solving: 7.0 Tool Calling: 10.0
#16	Gemini 3.1 Flash Lite Previewlow	Google	7.3	0.177	$0.020	3.36s
Lihat kartu model Total Tes: 16 Tes Salah: 5 Tingkat lulus per percobaan: 68.8% Tes tidak stabil: 0 Token output: 1,611 Token penalaran: 7,272 Waktu respons: rata-rata 3.36s · total 53.84s · maks 11.91s Jawaban salah: 4 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 10.0 Puzzle Solving: 10.0 Tool Calling: 10.0
#17	DeepSeek V3.2medium	DeepSeek	7.3	0.225	$0.025	39.48s
Lihat kartu model Total Tes: 16 Tes Salah: 5 Tingkat lulus per percobaan: 79.2% Tes tidak stabil: 3 Token output: 7,392 Token penalaran: 39,089 Waktu respons: rata-rata 39.48s · total 631.71s · maks 93.11s Jawaban salah: 3 Tidak mengikuti instruksi: 1 Kedaluwarsa: 1 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 10.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#18	Gemini 3 Flash Previewnone	Google	7.2	0.169	$0.019	1.75s
Lihat kartu model Total Tes: 16 Tes Salah: 5 Tingkat lulus per percobaan: 75.0% Tes tidak stabil: 2 Token output: 1,411 Token penalaran: 0 Waktu respons: rata-rata 1.75s · total 15.71s · maks 3.56s Jawaban salah: 5 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 7.0 General Intelligence: 10.0 Instructions following: 5.5 Puzzle Solving: 7.0 Tool Calling: 10.0
#19	MiMo-V2-Flashmedium	Xiaomi	7.2	0.316	$0.035	25.33s
Lihat kartu model Total Tes: 16 Tes Salah: 5 Tingkat lulus per percobaan: 72.9% Tes tidak stabil: 1 Token output: 11,613 Token penalaran: 106,714 Waktu respons: rata-rata 25.33s · total 253.33s · maks 96.01s Jawaban salah: 3 Kesalahan API: 1 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 9.7 Combined: 9.0 Data parsing and extraction: 5.5 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 10.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#20	Step 3.5 Flashmedium	Stepfun	7.4	0.000	$0.000	29.10s
Lihat kartu model Total Tes: 16 Tes Salah: 6 Tingkat lulus per percobaan: 68.8% Tes tidak stabil: 2 Token output: 71,452 Token penalaran: 155,147 Waktu respons: rata-rata 29.10s · total 290.96s · maks 170.45s Tidak mengikuti instruksi: 3 Jawaban salah: 3 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 10.0 Domain specific: 4.0 General Intelligence: 6.0 Instructions following: 9.0 Puzzle Solving: 4.0 Tool Calling: 10.0
#21	GPT-5.3 Chatnone	OpenAI	7.3	3.163	$0.317	5.96s
Lihat kartu model Total Tes: 16 Tes Salah: 6 Tingkat lulus per percobaan: 70.8% Tes tidak stabil: 3 Token output: 19,272 Token penalaran: 0 Waktu respons: rata-rata 5.96s · total 95.30s · maks 18.33s Jawaban salah: 4 Tidak mengikuti instruksi: 2 Anti-AI Tricks: 7.3 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 4.0 Instructions following: 9.0 Puzzle Solving: 10.0 Tool Calling: 10.0
#22	Gemini 3.1 Flash Lite Previewnone	Google	7.1	0.144	$0.015	1.33s
Lihat kartu model Total Tes: 16 Tes Salah: 6 Tingkat lulus per percobaan: 66.7% Tes tidak stabil: 1 Token output: 4,715 Token penalaran: 0 Waktu respons: rata-rata 1.33s · total 21.27s · maks 3.39s Jawaban salah: 4 Tidak mengikuti instruksi: 2 Anti-AI Tricks: 6.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 10.0 Puzzle Solving: 10.0 Tool Calling: 10.0
#23	Seed-2.0-Minimedium	Bytedance Seed	6.9	0.280	$0.028	65.09s
Lihat kartu model Total Tes: 16 Tes Salah: 6 Tingkat lulus per percobaan: 68.8% Tes tidak stabil: 2 Token output: 1,965 Token penalaran: 58,456 Waktu respons: rata-rata 65.09s · total 846.14s · maks 262.83s Kedaluwarsa: 4 Tidak mengikuti instruksi: 1 Jawaban salah: 1 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 6.0 Instructions following: 10.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#24	Qwen3.5-Flashmedium	Qwen	6.9	0.720	$0.072	70.81s
Lihat kartu model Total Tes: 16 Tes Salah: 6 Tingkat lulus per percobaan: 81.3% Tes tidak stabil: 5 Token output: 1,807 Token penalaran: 169,952 Waktu respons: rata-rata 70.81s · total 1132.90s · maks 234.29s Kedaluwarsa: 3 Kesalahan API: 1 Tidak mengikuti instruksi: 1 Jawaban salah: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 5.5 Domain specific: 4.0 General Intelligence: 5.0 Instructions following: 10.0 Puzzle Solving: 4.0 Tool Calling: 10.0
#25	Claude Sonnet 4.6none	Anthropic	6.8	2.504	$0.251	5.57s
Lihat kartu model Total Tes: 16 Tes Salah: 6 Tingkat lulus per percobaan: 66.7% Tes tidak stabil: 1 Token output: 6,895 Token penalaran: 0 Waktu respons: rata-rata 5.57s · total 50.12s · maks 23.84s Format tambahan: 3 Jawaban salah: 2 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 4.0 Combined: 9.0 Data parsing and extraction: 9.9 Domain specific: 7.0 General Intelligence: 5.0 Instructions following: 5.5 Puzzle Solving: 7.0 Tool Calling: 10.0
#26	Claude Opus 4.6medium	Anthropic	6.6	13.118	$1.312	22.86s
Lihat kartu model Total Tes: 16 Tes Salah: 6 Tingkat lulus per percobaan: 66.7% Tes tidak stabil: 2 Token output: 26,254 Token penalaran: 17,363 Waktu respons: rata-rata 22.86s · total 205.71s · maks 83.40s Format tambahan: 4 Jawaban salah: 2 Anti-AI Tricks: 4.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#27	GPT-5.2medium	OpenAI	6.5	3.125	$0.313	15.33s
Lihat kartu model Total Tes: 16 Tes Salah: 6 Tingkat lulus per percobaan: 75.0% Tes tidak stabil: 4 Token output: 2,220 Token penalaran: 16,811 Waktu respons: rata-rata 15.33s · total 138.01s · maks 77.80s Tidak mengikuti instruksi: 3 Tidak ada jawaban: 1 Kedaluwarsa: 1 Jawaban salah: 1 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 10.0 Instructions following: 9.5 Puzzle Solving: 7.0 Tool Calling: 10.0
#28	Kimi K2.5medium	Moonshot AI	6.4	2.171	$0.196	69.83s
Lihat kartu model Total Tes: 16 Tes Salah: 7 Tingkat lulus per percobaan: 72.9% Tes tidak stabil: 5 Token output: 38,453 Token penalaran: 72,496 Waktu respons: rata-rata 69.83s · total 628.45s · maks 137.29s Jawaban salah: 3 Tidak mengikuti instruksi: 2 Tidak ada jawaban: 1 Kedaluwarsa: 1 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 6.0 Instructions following: 10.0 Puzzle Solving: 4.0 Tool Calling: 10.0
#29	Qwen3.5 Plus 2026-02-15none	Qwen	6.2	0.172	$0.016	2.65s
Lihat kartu model Total Tes: 16 Tes Salah: 7 Tingkat lulus per percobaan: 58.3% Tes tidak stabil: 1 Token output: 2,015 Token penalaran: 0 Waktu respons: rata-rata 2.65s · total 26.52s · maks 6.65s Jawaban salah: 7 Anti-AI Tricks: 4.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 4.0 Instructions following: 10.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#30	Grok 4.1 Fastmedium	X AI	6.2	0.563	$0.051	26.35s
Lihat kartu model Total Tes: 16 Tes Salah: 7 Tingkat lulus per percobaan: 66.7% Tes tidak stabil: 4 Token output: 1,183 Token penalaran: 83,875 Waktu respons: rata-rata 26.35s · total 237.11s · maks 121.79s Tidak mengikuti instruksi: 3 Jawaban salah: 2 Tidak ada jawaban: 1 Kedaluwarsa: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 5.5 Puzzle Solving: 4.0 Tool Calling: 10.0
#31	GLM 5none	Z.ai	6.0	0.200	$0.018	4.03s
Lihat kartu model Total Tes: 16 Tes Salah: 7 Tingkat lulus per percobaan: 56.3% Tes tidak stabil: 0 Token output: 1,548 Token penalaran: 0 Waktu respons: rata-rata 4.03s · total 36.30s · maks 11.07s Jawaban salah: 7 Anti-AI Tricks: 4.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 7.0 Tool Calling: 10.0
#32	GPT-5 Minimedium	OpenAI	6.0	1.457	$0.117	25.14s
Lihat kartu model Total Tes: 16 Tes Salah: 8 Tingkat lulus per percobaan: 58.3% Tes tidak stabil: 2 Token output: 5,826 Token penalaran: 48,768 Waktu respons: rata-rata 25.14s · total 402.29s · maks 88.15s Tidak mengikuti instruksi: 4 Jawaban salah: 3 Kedaluwarsa: 1 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 4.0 Instructions following: 7.5 Puzzle Solving: 4.3 Tool Calling: 10.0
#33	Qwen3.5-35B-A3Bmedium	Qwen	5.5	4.251	$0.341	43.93s
Lihat kartu model Total Tes: 16 Tes Salah: 8 Tingkat lulus per percobaan: 77.1% Tes tidak stabil: 7 Token output: 5,495 Token penalaran: 169,266 Waktu respons: rata-rata 43.93s · total 702.85s · maks 106.00s Kedaluwarsa: 4 Jawaban salah: 2 Kesalahan API: 1 Tidak ada jawaban: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 5.5 Domain specific: 10.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 4.0 Tool Calling: 10.0
#34	DeepSeek V3.2none	DeepSeek	5.5	0.220	$0.016	12.86s
Lihat kartu model Total Tes: 16 Tes Salah: 9 Tingkat lulus per percobaan: 54.2% Tes tidak stabil: 3 Token output: 7,823 Token penalaran: 0 Waktu respons: rata-rata 12.86s · total 205.78s · maks 115.89s Jawaban salah: 6 Format tambahan: 2 common.failureReasons.invalid_tool_call: 1 Anti-AI Tricks: 10.0 Combined: 8.0 Data parsing and extraction: 5.4 Domain specific: 10.0 General Intelligence: 10.0 Instructions following: 10.0 Puzzle Solving: 7.7 Tool Calling: 10.0
#35	GPT-5 Nanomedium	OpenAI	5.5	0.856	$0.060	47.94s
Lihat kartu model Total Tes: 16 Tes Salah: 9 Tingkat lulus per percobaan: 60.4% Tes tidak stabil: 6 Token output: 4,386 Token penalaran: 142,080 Waktu respons: rata-rata 47.94s · total 431.47s · maks 204.02s Jawaban salah: 5 Tidak mengikuti instruksi: 3 Kedaluwarsa: 1 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 10.0 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 9.0 Puzzle Solving: 4.0 Tool Calling: 10.0
#36	Mercury 2medium	Inception	5.3	0.631	$0.045	2.36s
Lihat kartu model Total Tes: 16 Tes Salah: 9 Tingkat lulus per percobaan: 54.2% Tes tidak stabil: 3 Token output: 3,708 Token penalaran: 45,921 Waktu respons: rata-rata 2.36s · total 35.39s · maks 14.63s Jawaban salah: 5 Tidak mengikuti instruksi: 4 Anti-AI Tricks: 7.3 Combined: 10.0 Data parsing and extraction: 5.5 Domain specific: 10.0 General Intelligence: 4.0 Instructions following: 10.0 Puzzle Solving: 1.7 Tool Calling: 10.0
#37	Qwen3.5-Flashnone	Qwen	5.2	0.077	$0.006	3.54s
Lihat kartu model Total Tes: 16 Tes Salah: 9 Tingkat lulus per percobaan: 45.8% Tes tidak stabil: 1 Token output: 3,774 Token penalaran: 0 Waktu respons: rata-rata 3.54s · total 56.70s · maks 13.73s Jawaban salah: 8 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 2.3 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 7.0 General Intelligence: 10.0 Instructions following: 5.0 Puzzle Solving: 1.3 Tool Calling: 10.0
#38	gpt-oss-120bmedium	OpenAI	5.1	0.135	$0.010	16.65s
Lihat kartu model Total Tes: 16 Tes Salah: 9 Tingkat lulus per percobaan: 54.2% Tes tidak stabil: 5 Token output: 13,210 Token penalaran: 34,230 Waktu respons: rata-rata 16.65s · total 149.88s · maks 50.92s Jawaban salah: 5 Tidak mengikuti instruksi: 4 Anti-AI Tricks: 7.0 Combined: 10.0 Data parsing and extraction: 5.5 Domain specific: 10.0 General Intelligence: 3.0 Instructions following: 9.5 Puzzle Solving: 1.7 Tool Calling: 9.0
#39	Gemini 2.5 Flashnone	Google	5.2	0.191	$0.012	923ms
Lihat kartu model Total Tes: 16 Tes Salah: 10 Tingkat lulus per percobaan: 43.8% Tes tidak stabil: 2 Token output: 1,270 Token penalaran: 0 Waktu respons: rata-rata 923ms · total 14.78s · maks 4.39s Jawaban salah: 9 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 5.0 Instructions following: 9.0 Puzzle Solving: 4.7 Tool Calling: 10.0
#40	Qwen3.5-122B-A10Bnone	Qwen	5.0	0.333	$0.020	3.72s
Lihat kartu model Total Tes: 16 Tes Salah: 10 Tingkat lulus per percobaan: 39.6% Tes tidak stabil: 1 Token output: 2,679 Token penalaran: 0 Waktu respons: rata-rata 3.72s · total 59.46s · maks 46.00s Jawaban salah: 9 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 4.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 5.0 Instructions following: 4.5 Puzzle Solving: 4.0 Tool Calling: 10.0
#41	Qwen3.5-35B-A3Bnone	Qwen	4.7	0.237	$0.015	4.10s
Lihat kartu model Total Tes: 16 Tes Salah: 10 Tingkat lulus per percobaan: 50.0% Tes tidak stabil: 3 Token output: 3,756 Token penalaran: 0 Waktu respons: rata-rata 4.10s · total 65.62s · maks 47.43s Jawaban salah: 8 Tidak mengikuti instruksi: 2 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 7.0 General Intelligence: 6.0 Instructions following: 5.0 Puzzle Solving: 1.7 Tool Calling: 10.0
#42	GPT-5.4none	OpenAI	4.5	1.562	$0.094	1.48s
Lihat kartu model Total Tes: 16 Tes Salah: 10 Tingkat lulus per percobaan: 41.7% Tes tidak stabil: 2 Token output: 1,819 Token penalaran: 0 Waktu respons: rata-rata 1.48s · total 23.64s · maks 2.89s Jawaban salah: 9 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 5.5 Puzzle Solving: 4.0 Tool Calling: 10.0
#43	Qwen3.5-27Bnone	Qwen	4.9	0.302	$0.016	1.75s
Lihat kartu model Total Tes: 16 Tes Salah: 11 Tingkat lulus per percobaan: 37.5% Tes tidak stabil: 2 Token output: 3,161 Token penalaran: 0 Waktu respons: rata-rata 1.75s · total 28.05s · maks 9.39s Jawaban salah: 9 Tidak mengikuti instruksi: 2 Anti-AI Tricks: 4.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 5.0 Instructions following: 4.5 Puzzle Solving: 6.3 Tool Calling: 10.0
#44	MiniMax M2.5medium	Minimax	4.7	4.981	$0.250	43.03s
Lihat kartu model Total Tes: 16 Tes Salah: 11 Tingkat lulus per percobaan: 60.4% Tes tidak stabil: 9 Token output: 107,044 Token penalaran: 206,190 Waktu respons: rata-rata 43.03s · total 387.25s · maks 237.27s Jawaban salah: 5 Tidak mengikuti instruksi: 3 Kedaluwarsa: 2 common.failureReasons.invalid_tool_call: 1 Anti-AI Tricks: 9.3 Combined: 10.0 Data parsing and extraction: 10.0 Domain specific: 10.0 General Intelligence: 3.0 Instructions following: 8.0 Puzzle Solving: 4.0 Tool Calling: 10.0
#45	Trinity Large Previewnone	Arcee AI	4.2	0.000	$0.000	3.15s
Lihat kartu model Total Tes: 16 Tes Salah: 11 Tingkat lulus per percobaan: 33.3% Tes tidak stabil: 1 Token output: 1,837 Token penalaran: 0 Waktu respons: rata-rata 3.15s · total 50.46s · maks 8.91s Jawaban salah: 9 Tidak mengikuti instruksi: 2 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 3.5 Puzzle Solving: 4.0 Tool Calling: 10.0
#46	Kimi K2.5none	Moonshot AI	4.1	0.295	$0.015	11.91s
Lihat kartu model Total Tes: 16 Tes Salah: 11 Tingkat lulus per percobaan: 39.6% Tes tidak stabil: 3 Token output: 2,000 Token penalaran: 0 Waktu respons: rata-rata 11.91s · total 107.16s · maks 42.13s Jawaban salah: 11 Anti-AI Tricks: 2.7 Combined: 10.0 Data parsing and extraction: 5.4 Domain specific: 4.0 General Intelligence: 10.0 Instructions following: 5.5 Puzzle Solving: 10.0 Tool Calling: 10.0
#47	GPT-4o-mininone	OpenAI	4.0	0.114	$0.005	2.07s
Lihat kartu model Total Tes: 16 Tes Salah: 12 Tingkat lulus per percobaan: 25.0% Tes tidak stabil: 0 Token output: 1,594 Token penalaran: 0 Waktu respons: rata-rata 2.07s · total 18.60s · maks 7.58s Jawaban salah: 11 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 4.0 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 10.0 General Intelligence: 3.0 Instructions following: 4.5 Puzzle Solving: 2.3 Tool Calling: 10.0
#48	Qwen3 Coder Nextnone	Qwen	4.0	0.174	$0.007	11.68s
Lihat kartu model Total Tes: 16 Tes Salah: 12 Tingkat lulus per percobaan: 25.0% Tes tidak stabil: 0 Token output: 3,026 Token penalaran: 0 Waktu respons: rata-rata 11.68s · total 116.76s · maks 45.14s Jawaban salah: 10 Format tambahan: 1 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 2.3 Combined: 10.0 Data parsing and extraction: 5.4 Domain specific: 4.0 General Intelligence: 10.0 Instructions following: 4.5 Puzzle Solving: 1.3 Tool Calling: 10.0
#49	GLM 4.7 Flashnone	Z.ai	3.9	0.066	$0.003	2.99s
Lihat kartu model Total Tes: 16 Tes Salah: 12 Tingkat lulus per percobaan: 35.4% Tes tidak stabil: 3 Token output: 1,855 Token penalaran: 0 Waktu respons: rata-rata 2.99s · total 26.90s · maks 7.05s Jawaban salah: 9 Tidak mengikuti instruksi: 2 common.failureReasons.invalid_tool_call: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 5.4 Domain specific: 7.0 General Intelligence: 3.0 Instructions following: 5.5 Puzzle Solving: 3.7 Tool Calling: 10.0
#50	Mercury 2none	Inception	3.4	0.153	$0.007	596ms
Lihat kartu model Total Tes: 16 Tes Salah: 12 Tingkat lulus per percobaan: 31.3% Tes tidak stabil: 2 Token output: 1,303 Token penalaran: 0 Waktu respons: rata-rata 596ms · total 9.54s · maks 1.27s Jawaban salah: 11 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 5.5 Domain specific: 4.0 General Intelligence: 4.0 Instructions following: 5.5 Puzzle Solving: 10.0 Tool Calling: 10.0
#51	GLM 4.7 Flashmedium	Z.ai	3.1	1.040	$0.042	36.84s
Lihat kartu model Total Tes: 16 Tes Salah: 12 Tingkat lulus per percobaan: 41.7% Tes tidak stabil: 7 Token output: 38,682 Token penalaran: 64,952 Waktu respons: rata-rata 36.84s · total 331.58s · maks 174.55s Jawaban salah: 7 Tidak mengikuti instruksi: 2 Tidak ada jawaban: 2 common.failureReasons.invalid_tool_call: 1 Anti-AI Tricks: 4.0 Combined: 10.0 Data parsing and extraction: 5.0 Domain specific: 10.0 General Intelligence: 10.0 Instructions following: 5.0 Puzzle Solving: 10.0 Tool Calling: 10.0
#52	Qwen3 Coder Nextmedium	Qwen	3.5	0.230	$0.007	12.53s
Lihat kartu model Total Tes: 16 Tes Salah: 13 Tingkat lulus per percobaan: 27.1% Tes tidak stabil: 2 Token output: 2,935 Token penalaran: 0 Waktu respons: rata-rata 12.53s · total 125.32s · maks 81.80s Jawaban salah: 8 Tidak mengikuti instruksi: 5 Anti-AI Tricks: 1.3 Combined: 10.0 Data parsing and extraction: 5.4 Domain specific: 4.0 General Intelligence: 6.0 Instructions following: 4.5 Puzzle Solving: 10.0 Tool Calling: 10.0
#53	Grok 4.1 Fastnone	X AI	2.9	0.247	$0.008	1.90s
Lihat kartu model Total Tes: 16 Tes Salah: 13 Tingkat lulus per percobaan: 25.0% Tes tidak stabil: 2 Token output: 1,148 Token penalaran: 0 Waktu respons: rata-rata 1.90s · total 17.14s · maks 5.51s Jawaban salah: 11 Tidak mengikuti instruksi: 2 Anti-AI Tricks: 1.3 Combined: 10.0 Data parsing and extraction: 9.9 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 10.0 Puzzle Solving: 1.3 Tool Calling: 10.0
#54	MiMo-V2-Flashnone	Xiaomi	2.9	0.743	$0.023	2.97s
Lihat kartu model Total Tes: 16 Tes Salah: 13 Tingkat lulus per percobaan: 29.2% Tes tidak stabil: 4 Token output: 67,790 Token penalaran: 0 Waktu respons: rata-rata 2.97s · total 35.60s · maks 19.68s Jawaban salah: 10 Kesalahan API: 1 Format tambahan: 1 Tidak mengikuti instruksi: 1 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 10.0 Domain specific: 4.0 General Intelligence: 4.0 Instructions following: 5.5 Puzzle Solving: 10.0 Tool Calling: 10.0
#55	LFM2-24B-A2Bnone	Liquid	2.6	0.047	$0.001	811ms
Lihat kartu model Total Tes: 16 Tes Salah: 15 Tingkat lulus per percobaan: 14.6% Tes tidak stabil: 2 Token output: 1,185 Token penalaran: 0 Waktu respons: rata-rata 811ms · total 11.35s · maks 2.88s Jawaban salah: 9 Kesalahan API: 4 Tidak mengikuti instruksi: 2 Anti-AI Tricks: 10.0 Combined: 10.0 Data parsing and extraction: 10.0 Domain specific: 4.0 General Intelligence: 3.0 Instructions following: 4.5 Puzzle Solving: 3.3 Tool Calling: 10.0

Perbandingan Cepat