IBM watsonx AI models — pricing & benchmarks

Official siteDocsPricingAPI

IBM watsonx is a portfolio of AI products that accelerates the impact of generative AI in core workflows to drive productivity. Own models · Inference platform · OpenAI-compatible API · Compliance · Enterprise · Explainability · Granite · Regulated Industries

Intelligence vs Price

Best value among IBM watsonx models on this chart: GPT OSS 120B · Llama 3.2 3B Instruct. Hover any dot for full pricing, or click a creator in the legend to isolate.

IBM watsonx models

29 models, 28 with pricing

Model	Creator	Input Price, $	Output Price, $	Context	Max Output	Inference Providers	Intelligence	Coding
GPT OSS 120B	OpenAI	0.039	0.180	131K	131K	compare (20)	33.3#1	28.6#1
Llama 3.3 70B Instruct	Meta	0.100	0.200	131K	120K	compare (20)	14.5#2	10.7#2
Granite 4 Small H	IBM	0.060	0.250	20K	N/A	compare (1)	10.8#3	8.5#3
Mistral Large	Mistral AI	0.500	1.50	262K	16K	compare (8)	9.9#4	N/A
Llama 3.2 3B Instruct	Meta	0.015	0.020	131K	80K	compare (9)	9.7#5	N/A
Mistral Small	Mistral AI	0.060	0.180	262K	8K	compare (7)	9.0#7	N/A
Mistral Medium	Mistral AI	0.400	2.00	131K	64K	compare (4)	9.0#6	N/A
Granite 3.3 8B Instruct	IBM	0.030	0.200	8K	8K	compare (2)	7.0#8	3.4#4
Llama 3.2 1B Instruct	Meta	0.027	0.080	131K	60K	compare (5)	6.3#9	0.6#5
Allam 1 13B Instruct	SDAIA	1.80	1.80	8K	8K	compare (1)	N/A	N/A
Flan-T5 3B XL	Google	0.600	0.600	8K	8K	compare (1)	N/A	N/A
Granite 2 13B Chat	IBM	0.600	0.600	8K	8K	compare (1)	N/A	N/A
Granite 2 13B Instruct	IBM	0.600	0.600	8K	8K	compare (1)	N/A	N/A
Granite 3 8B Instruct	IBM	0.200	0.200	8K	1K	compare (1)	N/A	N/A
Granite Guardian 3.2 2B	IBM	0.100	0.100	8K	8K	compare (1)	N/A	N/A
Granite Guardian 3.3 8B	IBM	0.200	0.200	8K	8K	compare (1)	N/A	N/A
Granite TTM R2 1024 96	IBM	0.380	0.380	512	512	compare (1)	N/A	N/A
Granite TTM R2 1536 96	IBM	0.380	0.380	512	512	compare (1)	N/A	N/A
Granite TTM R2 512 96	IBM	0.380	0.380	512	512	compare (1)	N/A	N/A
Granite Vision 3.2 2B	IBM	0.100	0.100	8K	8K	compare (1)	N/A	N/A
Jais 13B Chat	Core42	0.500	2.00	8K	8K	compare (1)	N/A	N/A
Llama 3.2 11B Vision Instruct	Meta	0.015	0.025	131K	16K	compare (7)	N/A	N/A
Llama 3.2 90B Vision Instruct	Meta	0.900	0.900	128K	16K	compare (4)	N/A	N/A
Llama 4 Maverick 17B	Meta	0.350	1.40	128K	N/A	compare (1)	N/A	N/A
LlamaGuard 3 11B Vision	Meta	0.350	0.350	128K	N/A	compare (1)	N/A	N/A
Mistral Small 3.1 24B Instruct	Mistral AI	0.100	0.300	131K	N/A	compare (2)	N/A	N/A
mT0 13B XXL	BigScience	0.500	2.00	8K	8K	compare (1)	N/A	N/A
Pixtral 12B	Mistral AI	0.150	0.150	128K	4K	compare (3)	N/A	N/A
Whisper 3 Large Turbo	OpenAI	N/A	N/A	N/A	N/A	compare (2)	N/A	N/A