Provider pricing
IBM AI API pricing
IBM has 11 priced models in the public snapshot. For 1,000 input and 500 output tokens, Granite 4.0 H Micro has the lowest recorded request cost at $0.000073, with $0.017 input and $0.112 output per million tokens.
Best for
Regulated enterprises, financial services, and government teams needing sovereign-cloud LLMs with IP indemnity.
Strengths
- Open-source weights with enterprise IP indemnification
- Apache 2.0 license — fine-tune and self-host freely
- Strong RAG, summarization, and function calling for enterprise workflows
- Available on watsonx.ai with EU, US, and Japan data residency
Tracked IBM models
| Model | Input $/M | Output $/M | Context | Tier | Best for |
|---|---|---|---|---|---|
| Granite 3.3 2B Instruct | $0.100 | $0.100 | 128,000 | Budget | Ultra-cheap on-device/edge inference, classification, summarization on watsonx.ai |
| Granite 3.3 8B Instruct | $0.500 | $0.500 | 128,000 | Open Source | Enterprise RAG, summarization, function calling on watsonx.ai |
| Granite Code 8B Instruct | $0.200 | $0.200 | 128,000 | Budget | Code-specialized 8B for enterprise IDE integrations on watsonx.ai |
| Granite Vision 3.2 2B | $0.100 | $0.100 | 16,000 | Budget | Compact vision-language model for document and chart understanding |
| Granite 4 8B Instruct | $0.200 | $0.200 | 256,000 | Open Source | Hybrid Mamba-Transformer enterprise instruct on watsonx.ai |
Last updated: · Source: IBM pricing · Methodology
Frequently asked questions
How much does the IBM API cost?
IBM has 11 priced models in the public snapshot. For 1,000 input and 500 output tokens, Granite 4.0 H Micro has the lowest recorded request cost at $0.000073, with $0.017 input and $0.112 output per million tokens.
Which IBM model is cheapest?
Granite 4.0 H Micro is the lowest-cost recorded model for the 1,000-input/500-output example. Prices are $0.017 input and $0.112 output per million tokens; different workloads may change the ranking.
Does IBM have cached or batch prices?
Cached input: not separately recorded. Batch: not recorded in this dataset. Eligibility and cache-write charges depend on the provider; these discounts are not automatically stacked.
How often are provider prices verified?
The live pricing table synchronizes stored model records daily; public file snapshots update when the site is rebuilt. Synchronization is not independent provider verification. Snapshot updated 2026-10-31; only explicit per-model dates are verification records.