Value: quality per dollar

BenchLeader Index divided by the blended price per million tokens (3 input : 1 output). Higher means more quality for the money.

259 models · data as of 9 Sept 2026

1. 1qwen3-4b-instruct-25071390.0
2. 2Gemma 4 E4B1135.0
3. 3Granite 4.2 3B898.1
4. 4Qwen3.5-4B785.0
5. 5LFM2 24B A2B775.5
6. 6Sarvam 30B (high)770.0
7. 7HyperNova 60B 2605 (high, based on gpt-oss-120b)760.0
8. 8Llama 3.1 8B690.6
9. 9Nova Micro659.0
10. 10Granite 4.1 8B649.2
11. 11nvidia-nemotron-nano-9b-v2600.0
12. 12Gemma 3 12B IT590.5
13. 13gpt-oss-20b554.1
14. 14Sarvam 105B (high)553.4
15. 15nemotron-3-nano-30b-a3b552.3

SearchAll providersAI21 LabsAlibabaAllen Institute for AIAmazonAnt GroupAnt-groupAnthropicArcee AIBaiduCohereDeepSeekGoogleIBMInceptionInclusionAIKwaiKATLiquid AIMetaMicrosoftMiniMaxMistral AIMoonshot AIMultiverse ComputingNous ResearchNVIDIAOpenAIRekaReka AISarvamStepFunSwiss AI InitiativeThinking MachinesUpstagexAIXiaomiZhipu AIOpen weights259 of 259

#	Model	Index per $ ↓	Index ↕	Blended $/M ↕
1	qwen3-4b-instruct-2507Alibabaopen	1390.0	41.7	$0.030
2	Gemma 4 E4BGoogleopen	1135.0	45.4	$0.040
3	Granite 4.2 3BIBMopen	898.1	47.6	$0.053
4	Qwen3.5-4BAlibaba	785.0	47.1	$0.060
5	LFM2 24B A2BLiquid AIopen	775.5	41.1	$0.053
6	SSarvam 30B (high)Sarvamopen	770.0	38.5	$0.050
7	MHyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computingopen	760.0	49.4	$0.065
8	Llama 3.1 8BMetaopen	690.6	36.6	$0.053
9	Nova MicroAmazon	659.0	40.2	$0.061
10	Granite 4.1 8BIBMopen	649.2	40.9	$0.063
11	nvidia-nemotron-nano-9b-v2NVIDIA	600.0	42.0	$0.070
12	Gemma 3 12B ITNVIDIAopen	590.5	37.2	$0.063
13	gpt-oss-20bOpenAIopen	554.1	47.1	$0.085
14	SSarvam 105B (high)Sarvamopen	553.4	40.4	$0.073
15	nemotron-3-nano-30b-a3bNVIDIAopen	552.3	48.6	$0.088
16	Gemma 3 4B ITNVIDIAopen	547.6	34.5	$0.063
17	deepseek-r1-0528-qwen3-8bDeepSeekopen	542.6	36.9	$0.068
18	GLM-5.3-FlashZhipu AIopen	512.6	61.0	$0.119
19	Gemma 3n E4BGoogleopen	488.0	36.6	$0.075
20	Llama 3-8BMetaopen	479.5	35.0	$0.073
21	ILing 3.0 FlashInclusionAIopen	469.6	54.0	$0.115
22	Llama 3.2 1BMetaopen	459.2	32.6	$0.071
23	Granite 4.2 8BIBMopen	450.0	48.6	$0.108
24	Mistral Small 3.2Mistral AIopen	447.8	40.3	$0.090
25	Ministral 8B (latest)Mistral AIopen	382.0	38.2	$0.100
26	Ministral 3 3BMistral AIopen	376.0	37.6	$0.100
27	Amazon Nova LiteAmazon	375.2	39.4	$0.105
28	Granite 4.0 H SmallIBMopen	373.1	40.3	$0.108
29	GPT-5 nanoOpenAI	370.3	51.1	$0.138
30	Step 3.5 FlashStepFunopen	366.0	54.9	$0.150
31	Gemini 2.0 Flash-LiteGoogle	358.8	47.0	$0.131
32	XMiMo-V2-FlashXiaomiopen	356.0	53.4	$0.150
33	DeepSeek V4 FlashDeepSeekopen	353.0	59.3	$0.168
34	Gemma 4 12BGoogleopen	347.3	52.1	$0.150
35	XMiMo-V2.5Xiaomiopen	333.1	58.3	$0.175
36	Deepseek v4 Flash VisionDeepSeek	328.0	57.4	$0.175
37	GLM-4.7-FlashZhipu AIopen	326.8	50.0	$0.153
38	AOlmo 3 7B ThinkAllen Institute for AIopen	318.4	39.8	$0.125
39	Qwen PlusAlibaba	306.3	48.4	$0.158
40	Llama 3.2 3BMetaopen	305.8	36.7	$0.120
41	SApertus 8B InstructSwiss AI Initiativeopen	297.6	37.2	$0.125
42	Nemotron 3 Nano Omni 30B A3bNVIDIAopen	291.1	46.0	$0.158
43	Gemma 4 26B A4BGoogleopen	284.2	54.0	$0.190
44	Devstral SmallMistral AIopen	283.3	42.5	$0.150
45	Gemini 2.5 Flash Lite 09Google	278.9	48.8	$0.175
46	Phi-4-MiniNVIDIAopen	271.0	35.5	$0.131
47	Gemma-4-31B-ITNVIDIAopen	270.2	55.4	$0.205
48	Gemini 2.5 Flash-LiteGoogle	266.9	46.7	$0.175
49	Mistral Small 3.1Mistral AI	264.7	39.7	$0.150
50	Ministral 3 8BMistral AIopen	263.3	39.5	$0.150
51	Qwen3.8-Flash-NextAlibabaopen	257.8	59.3	$0.230
52	Gemini 2.0 FlashGoogle	257.7	45.1	$0.175
53	Mistral Small 3Mistral AI	254.7	38.2	$0.150
54	Qwen3.5-9BAlibabaopen	251.1	47.7	$0.190
55	Gemma 3 27BGoogleopen	242.1	39.7	$0.164
56	GPT-4.1 nanoOpenAI	217.7	38.1	$0.175
57	Qwen3 30B A3BAlibabaopen	211.2	45.4	$0.215
58	Ministral 3 14BMistral AIopen	202.5	40.5	$0.200
59	DeepSeek-R1-Distill-Qwen-14BDeepSeekopen	188.2	41.4	$0.220
60	ARing Flash 2.0Ant-group	182.3	45.2	$0.248

Show 100 more