Urambazaji
AI BENCHY
Advertise here

AI BENCHY Compare

MiniMax: MiniMax M3 vs StepFun: Step 3.7 Flash

Benchmark zimetengenezwa kutoka seti za majaribio za AI BENCHY tarehe: 2026-06-01

Kipimo MiniMax M3 MiniMax M3 medium Toleo: 2026-06-01 Step 3.7 Flash Step 3.7 Flash high Toleo: 2026-05-29
Alama 7.3 7.1
Nafasi #65 #74
Uaminifu 9.6 10.0
Uthabiti 8.4 8.2
Majaribio sahihi
Kiwango cha kupita kwa kila jaribio 68.3% 65.0%
Majaribio yasiyo thabiti 6 4
Jumla ya uendeshaji 60 60
Gharama kwa matokeo 1.083 8.723
Jumla ya gharama $0.120 $0.960
Bei ya ingizo $0.300 / 1M $0.200 / 1M
Bei ya toleo $1.200 / 1M $1.150 / 1M
Tokeni za matokeo 46,884 828,084
Tokeni za hoja 85,935 0
Muda wa majibu (wastani) 68.44s 49.43s
Muda wa majibu (upeo) 431.03s 192.75s
Muda wa majibu (jumla) 1300.32s 988.58s

Modeli bora kwa alama

Alama dhidi ya gharama ya jumla

Muda wa majibu (wastani)

Alama vs Muda wa majibu (wastani)

Jumla ya tokeni za matokeo

Alama vs Jumla ya tokeni za matokeo

Mgawanyo wa kategoria

Mbinu za kupinga AI Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 5.5 3.7 66.7% 3 14.95s 874 3,414
Step 3.7 Flash 10.0 10.0 100.0% 0 13.40s 42,656 0
Uandishi wa msimbo Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 7.5 10.0 66.7% 1 185.58s 4,071 26,059
Step 3.7 Flash 3.2 4.6 16.7% 1 126.82s 164,069 0
Mchanganyiko Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 10.0 10.0 100.0% 0 65.30s 1,306 6,253
Step 3.7 Flash 10.0 10.0 100.0% 0 13.01s 8,802 0
Uchanganuzi na uchimbaji wa data Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 10.0 10.0 100.0% 0 14.92s 514 3,164
Step 3.7 Flash 10.0 10.0 100.0% 0 14.72s 23,113 0
Mahususi kwa domeni Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 6.0 10.0 44.4% 1 233.13s 16,254 19,070
Step 3.7 Flash 4.1 4.4 44.5% 2 149.64s 410,502 0
Akili ya jumla Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 5.1 3.4 33.3% 1 33.25s 2,487 2,523
Step 3.7 Flash 5.5 10.0 0.0% 0 4.17s 2,862 0
Ufuataji wa maagizo Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 9.8 10.0 100.0% 0 6.14s 103 920
Step 3.7 Flash 9.8 10.0 100.0% 0 1.52s 2,010 0
Utatuzi wa mafumbo Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 7.9 9.9 66.7% 0 49.91s 11,946 13,761
Step 3.7 Flash 5.3 7.2 44.4% 1 10.22s 25,422 0
Mwito wa zana Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 10.0 10.0 100.0% 0 11.91s 281 555
Step 3.7 Flash 10.0 10.0 100.0% 0 2.79s 1,172 0
Maarifa ya jumla Alama Uthabiti Kiwango cha kupita kwa kila jaribio Majaribio yasiyo thabiti Majaribio sahihi Muda wa majibu (wastani) Tokeni za matokeo Tokeni za hoja
MiniMax M3 3.0 10.0 0.0% 0 100.80s 9,048 10,216
Step 3.7 Flash 3.0 10.0 0.0% 0 149.34s 147,476 0

Ulinganisho wa haraka

Badilisha jozi ya ulinganisho

Claude Opus 4.8nonevsMiniMax M3mediumQwen3.6 Max PreviewnonevsStep 3.7 FlashhighGPT-5 MinimediumvsStep 3.7 FlashhighGPT-5.4 NanomediumvsStep 3.7 FlashhighStep 3.7 FlashhighvsMiMo-V2-FlashmediumRing-2.6-1TnonevsMiniMax M3mediumSeed-2.0-MinimediumvsStep 3.7 FlashhighMiniMax M3mediumvsStep 3.7 FlashlowClaude Opus 4.6mediumvsStep 3.7 FlashhighClaude Sonnet 4.6nonevsStep 3.7 FlashhighMiniMax M3mediumvsGPT-5.3 ChatnoneRing-2.6-1TnonevsStep 3.7 Flashhigh