Model yang Dibandingkan

Model yang direkomendasikan Qwen3.8 27B

It offers the best overall trade-off: a competitive score (5.5), lower cost than model lain dalam perbandingan ini, and balanced response time.

Perbandingan terperinci

Metrik	Qwen3.8 27B Qwen3.8 27B high Rilis: 2026-08-14	Qwen3.8 27B Qwen3.8 27B medium Rilis: 2026-08-14	Qwen3.8 27B Qwen3.8 27B low Rilis: 2026-08-14	Qwen3.8 27B Qwen3.8 27B none Rilis: 2026-08-14

Metrik	Qwen3.8 27B Qwen3.8 27B high Rilis: 2026-08-14	Qwen3.8 27B Qwen3.8 27B medium Rilis: 2026-08-14	Qwen3.8 27B Qwen3.8 27B low Rilis: 2026-08-14	Qwen3.8 27B Qwen3.8 27B none Rilis: 2026-08-14
Skor	8.4	7.8	7.9	5.5
Peringkat	#42	#68	#61	#214
Keandalan	9.6	9.6	10.0	10.0
Konsistensi	9.2	9.7	10.0	10.0
Percobaan	66/66	66/66	66/66	66/66
Tes benar
Tingkat lulus per percobaan	77.3%	66.7%	68.2%	40.9%
Tes tidak stabil	2	1	0	0
Total Run	66	66	66	66
Biaya per hasil	~1.792	~0.409	~0.473	~0.063
Total Biaya	~$0.287	~$0.058	~$0.071	~$0.006
Harga input	T/A	T/A	T/A	T/A
Harga output	T/A	T/A	T/A	T/A
Total token input	100,179	98,266	99,705	122,463
Token output	904	1,861	1,545	11,445
Token penalaran	655,731	134,301	167,784	0
Waktu respons (rata-rata)	167.89s	33.05s	39.11s	3.12s
Waktu respons (maks)	640.85s	214.63s	376.04s	44.76s
Waktu respons (total)	3693.54s	694.04s	860.51s	68.69s
Parameter	27.3B	27.3B	27.3B	27.3B
Ketersediaan	Bobot tersedia	Bobot tersedia	Bobot tersedia	Bobot tersedia

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

high

medium

low

none

Kategori:

Trik anti-AI	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	10.0	10.0	100.0%	8.10s	1,128	114	5,319
Qwen3.8 27B	10.0	10.0	100.0%	2.93s	672	228	1,842
Qwen3.8 27B	10.0	10.0	100.0%	3.02s	984	183	2,016
Qwen3.8 27B	6.5	10.0	50.0%	828ms	696	267	0

Pemrograman	Skor	Konsistensi	Tingkat lulus per percobaan	Tes tidak stabil	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	8.1	7.0	88.9%	1	248.62s	8,235	346	143,335
Qwen3.8 27B	10.0	10.0	100.0%	0	40.50s	7,893	585	26,808
Qwen3.8 27B	7.7	10.0	66.7%	0	140.92s	8,127	585	86,694
Qwen3.8 27B	5.5	10.0	33.3%	0	1.19s	7,911	408	0

Gabungan	Skor	Konsistensi	Tingkat lulus per percobaan	Tes tidak stabil	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	10.0	10.0	100.0%	0	185.57s	68,997	66	72,556
Qwen3.8 27B	8.7	6.9	83.3%	1	124.48s	77,767	70	51,705
Qwen3.8 27B	10.0	10.0	100.0%	0	90.63s	69,687	66	37,791
Qwen3.8 27B	3.0	10.0	0.0%	0	24.10s	95,181	9,588	0

Parsing dan ekstraksi data	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	10.0	10.0	100.0%	34.85s	9,675	123	10,320
Qwen3.8 27B	6.5	10.0	50.0%	8.76s	9,039	144	2,271
Qwen3.8 27B	10.0	10.0	100.0%	8.03s	9,087	150	1,962
Qwen3.8 27B	6.5	10.0	50.0%	1.14s	7,788	246	0

Spesifik domain	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	5.3	10.0	33.3%	526.74s	975	6	234,555
Qwen3.8 27B	5.3	10.0	33.3%	76.98s	780	12	41,184
Qwen3.8 27B	5.3	10.0	33.3%	60.78s	1,014	12	31,872
Qwen3.8 27B	7.7	10.0	66.7%	552ms	798	27	0

Kecerdasan umum	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	5.0	10.0	0.0%	8.45s	630	63	1,191
Qwen3.8 27B	5.0	10.0	0.0%	9.20s	516	204	1,299
Qwen3.8 27B	5.0	10.0	0.0%	5.37s	594	132	729
Qwen3.8 27B	4.2	9.9	0.0%	1.31s	522	126	0

Kepatuhan instruksi	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	9.8	10.0	100.0%	8.24s	927	42	2,871
Qwen3.8 27B	10.0	10.0	100.0%	2.54s	699	48	984
Qwen3.8 27B	10.0	10.0	100.0%	2.43s	855	48	915
Qwen3.8 27B	6.3	10.0	50.0%	582ms	711	72	0

Pemecahan teka-teki	Skor	Konsistensi	Tingkat lulus per percobaan	Tes tidak stabil	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	7.6	7.2	77.8%	1	199.90s	1,038	120	138,427
Qwen3.8 27B	8.3	10.0	66.7%	0	12.62s	696	444	6,237
Qwen3.8 27B	7.7	10.0	66.7%	0	4.91s	930	240	2,382
Qwen3.8 27B	6.0	10.0	33.3%	0	1.25s	714	411	0

Pemanggilan alat	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	10.0	10.0	100.0%	5.62s	8,256	21	912
Qwen3.8 27B	3.0	10.0	0.0%	0ms	0	0	0
Qwen3.8 27B	3.0	10.0	0.0%	8.42s	8,145	15	1,350
Qwen3.8 27B	10.0	10.0	100.0%	2.75s	7,932	273	0

Pengetahuan umum	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Qwen3.8 27B	3.0	10.0	0.0%	263.98s	318	3	46,245
Qwen3.8 27B	3.0	10.0	0.0%	11.29s	204	126	1,971
Qwen3.8 27B	3.0	10.0	0.0%	12.64s	282	114	2,073
Qwen3.8 27B	3.0	10.0	0.0%	703ms	210	27	0

Ganti Pasangan Perbandingan