AI BENCHY Compare

IBM: Granite 4.1 8B vs Mistral: Mistral Small 4

Benchmark zimetengenezwa kutoka seti za majaribio za AI BENCHY tarehe: 2026-05-01

Kipimo	Granite 4.1 8B Granite 4.1 8B none Toleo: 2026-05-01	Mistral Small 4 Mistral Small 4 medium Toleo: 2026-03-16

Kipimo	Granite 4.1 8B Granite 4.1 8B none Toleo: 2026-05-01	Mistral Small 4 Mistral Small 4 medium Toleo: 2026-03-16
Alama	4.3	5.7
Nafasi	#135	#99
Uaminifu	10.0	Haipo
Uthabiti	10.0	6.8
Majaribio sahihi
Kiwango cha kupita kwa kila jaribio	11.1%	50.0%
Majaribio yasiyo thabiti	0	7
Jumla ya uendeshaji	54	54
Gharama kwa matokeo	0.110	0.674
Jumla ya gharama	$0.003	$0.034
Bei ya ingizo	$0.050 / 1M	$0.150 / 1M
Bei ya toleo	$0.100 / 1M	$0.600 / 1M
Tokeni za matokeo	2,712	15,084
Tokeni za hoja	0	39,408
Muda wa majibu (wastani)	767ms	5.64s
Muda wa majibu (upeo)	2.17s	30.49s
Muda wa majibu (jumla)	13.81s	101.52s

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
Granite 4.1 8B	4.9	10.0	25.0%	0		844ms	903	0
Mistral Small 4	5.6	3.8	66.7%	3		2.67s	4,055	4,778

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
Granite 4.1 8B	6.9	10.0	0.0%	0		1.07s	339	0
Mistral Small 4	6.7	3.5	66.7%	1		30.49s	2,796	11,296

Mchanganyiko	Alama	Uthabiti	Kiwango cha kupita kwa kila jaribio	Majaribio yasiyo thabiti	Majaribio sahihi	Muda wa majibu (wastani)	Tokeni za matokeo	Tokeni za hoja
Granite 4.1 8B	3.0	10.0	0.0%	0		1.88s	396	0
Mistral Small 4	3.0	10.0	0.0%	0		25.25s	2,612	10,700

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
Granite 4.1 8B	3.0	10.0	0.0%	0		575ms	195	0
Mistral Small 4	7.3	5.9	83.3%	1		1.23s	335	723

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
Granite 4.1 8B	3.0	10.0	0.0%	0		357ms	24	0
Mistral Small 4	5.3	7.2	44.4%	1		6.11s	2,621	6,904

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
Granite 4.1 8B	4.0	10.0	0.0%	0		499ms	115	0
Mistral Small 4	4.8	10.0	0.0%	0		2.05s	821	828

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
Granite 4.1 8B	3.6	9.9	0.0%	0		344ms	66	0
Mistral Small 4	7.3	5.8	83.3%	1		1.38s	540	1,031

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
Granite 4.1 8B	3.7	10.0	0.0%	0		635ms	431	0
Mistral Small 4	3.4	9.7	0.0%	0		2.00s	983	2,338

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
Granite 4.1 8B	10.0	10.0	100.0%	0		2.17s	243	0
Mistral Small 4	10.0	10.0	100.0%	0		3.50s	321	810

Ulinganisho wa haraka

Badilisha jozi ya ulinganisho

Mistral Small 4mediumvsGLM 4.7 Flashnone Granite 4.1 8BnonevsQwen3.5-9Bmedium Mistral Small 4mediumvsKimi K2.5none Mistral Small 4mediumvsGLM 5.1none Mistral Small 4mediumvsQwen3.6 Flashnone Mistral Small 4mediumvsGLM 5 Turbonone Mistral Small 4mediumvsMiMo-V2.5-Pronone Mistral Small 4mediumvsQwen3.5 Plus 2026-04-20none Mistral Small 4mediumvsKimi K2.6none Mistral Small 4mediumvsGPT-5.4none Mistral Small 4mediumvsQwen3.5-122B-A10Bnone Mistral Small 4mediumvsGrok 4.20none