Model yang Dibandingkan

Perbandingan benchmark Nemotron 3 Super (medium) vs Qwen3.5-122B-A10B (medium) vs Elephant Alpha (medium) vs gpt-oss-120b (medium): Qwen3.5-122B-A10B (medium) unggul pada Skor dengan 7.1. Qwen3.5-122B-A10B (medium) unggul pada Keandalan dengan 10.0. Elephant Alpha (medium) memiliki Total Biaya terendah di $0.000. Elephant Alpha (medium) paling cepat di 1.27s.

Benchmark dihasilkan dari suite pengujian AI BENCHY pada: 2026-09-04

Peringkat: #217
Total token output: 115,574
Waktu respons (rata-rata): 52.31s
Total Biaya: $0.050

Peringkat: #127
Total token output: 487,519
Waktu respons (rata-rata): 65.67s
Total Biaya: $1.207

Peringkat: #295
Total token output: 2,596
Waktu respons (rata-rata): 1.27s
Total Biaya: $0.000

Peringkat: #189
Total token output: 98,282
Waktu respons (rata-rata): 20.81s
Total Biaya: $0.020

Model yang direkomendasikan gpt-oss-120b (medium)

It offers the best overall trade-off: a competitive score (6.1), lower cost than model lain dalam perbandingan ini, and balanced response time.

Perbandingan terperinci

Metrik	Nemotron 3 Super Nemotron 3 Super medium Rilis: 2026-03-11 Tersedia gratis	Qwen3.5-122B-A10B Qwen3.5-122B-A10B medium Rilis: 2026-02-24	Elephant Alpha Elephant Alpha medium Rilis: 2026-04-14	gpt-oss-120b gpt-oss-120b medium Rilis: 2025-08-05

Metrik	Nemotron 3 Super Nemotron 3 Super medium Rilis: 2026-03-11 Tersedia gratis	Qwen3.5-122B-A10B Qwen3.5-122B-A10B medium Rilis: 2026-02-24	Elephant Alpha Elephant Alpha medium Rilis: 2026-04-14	gpt-oss-120b gpt-oss-120b medium Rilis: 2025-08-05
Skor	5.7	7.1	4.3	6.1
Peringkat	#217	#127	#295	#189
Keandalan	9.1	10.0	T/A	10.0
Konsistensi	8.9	8.5	9.2	8.0
Percobaan	66/66	66/66	63/66	66/66
Tes benar
Tingkat lulus per percobaan	40.9%	71.2%	28.8%	50.0%
Tes tidak stabil	3	4	1	5
Total Run	66	66	63	66
Biaya per hasil	0.004	8.446	0.000	0.224
Total Biaya	$0.050	$1.207	$0.000	$0.020
Harga input	$0.085 / 1M	$0.290 / 1M	$0.000 / 1M	$0.037 / 1M
Harga output	$0.400 / 1M	$2.400 / 1M	$0.000 / 1M	$0.170 / 1M
Total token input	81,438	124,780	33,744	108,767
Token output	17,674	47,694	2,596	29,378
Token penalaran	97,900	439,825	0	68,904
Waktu respons (rata-rata)	52.31s	65.67s	1.27s	20.81s
Waktu respons (maks)	431.98s	519.30s	3.70s	68.16s
Waktu respons (total)	1046.21s	1444.68s	22.82s	332.88s
Parameter	120B total (12B aktif)	122B total (10B aktif)	104B total (7.4B aktif)	117B total (5.1B aktif)
Ketersediaan	Bobot tersedia	Sumber terbuka	Sumber terbuka	Sumber terbuka

Showcase generasi model

Hamster playing table tennis

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

#217 Nemotron 3 Super

medium

Biaya: $0.000
Waktu: 272.6s
Token: 5,296 tok

#127 Qwen3.5-122B-A10B

medium

Biaya: $0.019
Waktu: 48.7s
Token: 6,034 tok

#295 Elephant Alpha

medium

Elephant Alpha was a stealth model revealed on April 21st as Ling-2.6-flash. Find it here: https://openrouter.ai/inclusionai/ling-2.6-flash:free

Biaya: $0.000
Waktu: 0.1s
Token: 0 tok

#189 gpt-oss-120b

medium

Biaya: $0.001
Waktu: 26.7s
Token: 555 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

Kategori:

Trik anti-AI	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	8.3	10.0	75.0%	7.85s	686	748	1,305
Qwen3.5-122B-A10B	10.0	10.0	100.0%	9.75s	672	269	16,835
Elephant Alpha	6.6	10.0	50.0%	1.19s	726	815	0
gpt-oss-120b	6.7	9.9	50.0%	10.21s	1,314	3,518	2,177

Pemrograman	Skor	Konsistensi	Tingkat lulus per percobaan	Tes tidak stabil	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	3.1	10.0	0.0%	0	147.32s	2,275	797	4,424
Qwen3.5-122B-A10B	6.0	7.2	55.6%	1	114.48s	7,630	8,057	82,578
Elephant Alpha	3.7	7.8	11.1%	1	1.30s	813	365	0
gpt-oss-120b	5.9	7.0	55.6%	1	38.37s	7,782	3,365	11,973

Gabungan	Skor	Konsistensi	Tingkat lulus per percobaan	Tes tidak stabil	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	6.4	5.8	66.7%	1	259.89s	59,463	5,542	75,095
Qwen3.5-122B-A10B	6.4	5.8	66.7%	1	313.55s	97,886	18,373	203,450
Elephant Alpha	1.5	5.0	0.0%	0	3.70s	14,046	562	0
gpt-oss-120b	6.5	10.0	50.0%	0	23.96s	81,198	10,453	22,883

Parsing dan ekstraksi data	Skor	Konsistensi	Tingkat lulus per percobaan	Tes tidak stabil	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	10.0	10.0	100.0%	0	18.16s	7,944	877	2,607
Qwen3.5-122B-A10B	10.0	10.0	100.0%	0	23.41s	7,782	270	16,558
Elephant Alpha	6.5	10.0	50.0%	0	979ms	8,004	246	0
gpt-oss-120b	6.4	5.9	66.7%	1	1.98s	7,476	241	1,114

Spesifik domain	Skor	Konsistensi	Tingkat lulus per percobaan	Tes tidak stabil	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	2.9	4.4	22.2%	2	19.35s	465	4,558	5,119
Qwen3.5-122B-A10B	2.9	7.2	11.1%	1	74.43s	780	19,154	61,573
Elephant Alpha	3.0	10.0	0.0%	0	925ms	810	24	0
gpt-oss-120b	2.9	4.4	22.2%	2	27.55s	1,286	6,390	21,466

Kecerdasan umum	Skor	Konsistensi	Tingkat lulus per percobaan	Tes tidak stabil	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	4.1	10.0	0.0%	0	6.91s	492	105	363
Qwen3.5-122B-A10B	3.4	2.2	33.3%	1	34.11s	344	66	7,592
Elephant Alpha	4.3	10.0	0.0%	0	920ms	540	105	0
gpt-oss-120b	4.3	10.0	0.0%	0	7.90s	659	107	387

Kepatuhan instruksi	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	7.3	10.0	50.0%	6.97s	723	956	2,383
Qwen3.5-122B-A10B	10.0	10.0	100.0%	9.88s	593	77	7,372
Elephant Alpha	9.8	10.0	100.0%	987ms	732	82	0
gpt-oss-120b	9.9	10.0	100.0%	7.63s	1,036	126	1,799

Pemecahan teka-teki	Skor	Konsistensi	Tingkat lulus per percobaan	Tes tidak stabil	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	3.0	10.0	0.0%	0	3.15s	708	570	1,322
Qwen3.5-122B-A10B	10.0	10.0	100.0%	0	17.89s	696	284	27,575
Elephant Alpha	5.3	10.0	33.3%	0	868ms	729	166	0
gpt-oss-120b	5.3	7.2	44.4%	1	21.71s	1,190	1,790	2,264

Pemanggilan alat	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	10.0	10.0	100.0%	39.75s	8,544	270	1,969
Qwen3.5-122B-A10B	10.0	10.0	100.0%	4.60s	8,193	322	1,226
Elephant Alpha	3.0	10.0	0.0%	2.83s	7,344	231	0
gpt-oss-120b	9.8	10.0	100.0%	6.91s	6,514	287	1,083

Pengetahuan umum	Skor	Konsistensi	Tingkat lulus per percobaan	Waktu respons (rata-rata)	Token input	Token output	Token penalaran
Nemotron 3 Super	3.0	10.0	0.0%	55.32s	138	3,251	3,313
Qwen3.5-122B-A10B	3.0	10.0	0.0%	52.87s	204	822	15,066
Elephant Alpha	3.0	10.0	0.0%	0ms	0	0	0
gpt-oss-120b	3.0	10.0	0.0%	26.51s	312	3,101	3,758

Perbandingan Cepat

Ganti Pasangan Perbandingan

gpt-oss-120bmediumvsQwen3.7 Flashnone Gemini 3.1 Flash Litenonevsgpt-oss-120bmedium gpt-oss-120bmediumvsInklinglow Nemotron 3 SupermediumTersedia gratisvsGLM 5none Muse Glimmer 30BlowvsQwen3.5-122B-A10Bmedium gpt-oss-120bmediumvsQwen3.6 Flashnone Qwen3.5-122B-A10BmediumvsStep 3.7 Flashlow Gemini 3.1 Flash Liteminimalvsgpt-oss-120bmedium Nemotron 3 SupermediumTersedia gratisvsInkling Smalllow Gemma 4 31BnoneTersedia gratisvsgpt-oss-120bmedium Claude Sonnet 5nonevsgpt-oss-120bmedium gpt-oss-120bmediumvsQwen3.5 Plus 2026-04-20none