SambaNova AI models — pricing & benchmarks

Official siteDocsPricingAPI

Discover SambaNova - the complete AI platform delivering the fastest AI inference, fine-tuning, and scalable solutions for agentic AI easily integrated into existing data center infrastructures. Inference platform · OpenAI-compatible API · Enterprise · Fast Inference · Open Weight · Rdu

Intelligence vs Price

Best value among SambaNova models on this chart: MiniMax M2.7 · DeepSeek V3.1 · QwQ 32B (and 3 more on the dashed frontier). Hover any dot for full pricing, or click a creator in the legend to isolate.

SambaNova models

15 models, 15 with pricing

Model	Creator	Input Price, $	Output Price, $	Context	Max Output	Inference Providers	Intelligence	Coding
MiniMax M2.7	MiniMax	0.279	1.20	205K	131K	compare (4)	49.6#1	41.9#1
DeepSeek V3.1	DeepSeek	0.135	0.500	164K	66K	compare (14)	28.1#2	28.4#2
DeepSeek V3 324	DeepSeek	0.200	0.400	164K	16K	compare (13)	22.3#3	22.0#3
QwQ 32B	Alibaba	0.150	0.200	131K	16K	compare (8)	19.7#4	N/A
DeepSeek R1	DeepSeek	0.280	0.400	164K	66K	compare (14)	18.8#5	15.9#4
Llama 3.1 405B Instruct	Meta	0.120	0.300	131K	16K	compare (11)	17.4#6	14.5#5
DeepSeek R1 Distill Llama 70B	DeepSeek	0.200	0.375	131K	16K	compare (11)	16.0#7	11.4#6
Llama 3.3 70B Instruct	Meta	0.100	0.200	131K	120K	compare (20)	14.5#8	10.7#7
Llama 3.1 8B Instruct	Meta	0.020	0.030	200K	128K	compare (20)	11.8#9	4.9#8
Llama 3.2 3B Instruct	Meta	0.015	0.020	131K	80K	compare (9)	9.7#10	N/A
Llama 3.2 1B Instruct	Meta	0.027	0.080	131K	60K	compare (5)	6.3#11	0.6#9
Llama 4 17B Maverick Instruct	Meta	0.050	0.100	1.0M	16K	compare (9)	N/A	N/A
Llama 4 17B Scout Instruct	Meta	0.050	0.100	10.0M	16K	compare (11)	N/A	N/A
LlamaGuard 3 8B	Meta	0.020	0.030	131K	16K	compare (5)	N/A	N/A
Qwen3 32B	Alibaba	0.050	0.100	131K	41K	compare (15)	N/A	N/A