---
title: "IBM watsonx.ai API pricing per token: 9 models | BroadRouter"
description: "IBM watsonx.ai API pricing per 1M tokens for 9 models, from USD 0.0636 input, cheapest-host checks and maker price comparison."
url: https://broadrouter.com/llm-api-pricing/providers/ibm-watsonx/
lang: en
updated: 2026-11-15
publisher: BroadRouter (a BroadNet site); API by BroadNet Inference, https://inference.broadnet.ai
---

# IBM watsonx.ai API pricing

Updated 2026-11-15

IBM watsonx.ai lists standard per-token prices for 9 models. The cheapest is Granite 4.0 H Small at USD 0.0636 input and USD 0.265 output per 1M tokens; the dearest is Mistral Medium 3 at USD 3.18 / USD 9.50. It publishes no cached-input prices, and lists no per-model batch, flex or priority prices on the pages we read. Checked 2026-11-15.

IBM watsonx.ai is a model maker: 2 of the 9 models here are its own. By maker: Meta 3, Mistral AI 3, IBM 2, OpenAI 1. 9 prices, each read from [IBM watsonx.ai's own pricing page](https://www.ibm.com/products/watsonx-ai/pricing), checked 2026-11-15. BroadRouter is BroadNet's own site and does not sell IBM watsonx.ai's services. Automated re-check 2026-11-16: 9 of 9 prices still matched the provider's page.

Models with a standard price

9

Median input / output

USD 0.371 / USD 0.636

Cached-input prices

0 of 9

Cheapest host for

0 of 7 shared models

## Which models is IBM watsonx.ai the cheapest provider for?

None. Every one of the 7 models that IBM watsonx.ai shares with other providers is cheaper elsewhere at the standard tier. The closest is [Mistral Small 3.1 24B Instruct](https://broadrouter.com/llm-api-pricing/mistral-small-3.1-24b-instruct/), where IBM watsonx.ai ranks 2 of 4. It is the dearest provider for 3.

## How does IBM watsonx.ai price the most widely hosted models?

The 6 models IBM watsonx.ai sells that the most other providers also sell, with IBM watsonx.ai's rank among them on the 3:1 blend.

IBM watsonx.ai on the most widely hosted models
| Model | IBM watsonx.ai | Rank | Cheapest |
| --- | --- | --- | --- |
| [OpenAI GPT OSS 120B](https://broadrouter.com/llm-api-pricing/gpt-oss-120b/) | USD 0.159 / USD 0.636 | 18 of 21 | CoreWeave Forge USD 0.03 / USD 0.17 |
| [Llama 3.3 70B](https://broadrouter.com/llm-api-pricing/llama-3.3-70b-instruct/) | USD 0.7526 / USD 0.7526 | 12 of 14 | DeepInfra USD 0.10 / USD 0.32 |
| [Llama 4 Maverick 17B 128E Instruct FP8](https://broadrouter.com/llm-api-pricing/llama-4-maverick/) | USD 0.371 / USD 1.484 | 7 of 7 | GMI Cloud USD 0.25 / USD 0.80 |
| [Mistral Small 3.1 24B Instruct](https://broadrouter.com/llm-api-pricing/mistral-small-3.1-24b-instruct/) | USD 0.106 / USD 0.318 | 2 of 4 | Google Vertex AI USD 0.10 / USD 0.30 |
| [Mistral Large 3](https://broadrouter.com/llm-api-pricing/mistral-large-2512/) | USD 0.636 / USD 1.908 | 2 of 3 | Amazon Bedrock USD 0.50 / USD 1.50 |
| [Llama 3.2 11B Vision Instruct](https://broadrouter.com/llm-api-pricing/llama-3.2-11b-vision-instruct/) | USD 0.371 / USD 0.371 | 3 of 3 | Amazon Bedrock USD 0.16 / USD 0.16 |

## Which model makers does IBM watsonx.ai sell?

IBM watsonx.ai prices models from 4 makers. Meta leads with 3 models. Standard prices, USD per 1M input / output tokens.

IBM watsonx.ai models by maker
| Maker | Models | Cheapest model | Median input / output |
| --- | --- | --- | --- |
| Meta | 3 | Llama 3.2 11B Vision Instruct USD 0.371 / USD 0.371 | USD 0.371 / USD 0.7526 |
| Mistral AI | 3 | Mistral Small 3.1 24B Instruct USD 0.106 / USD 0.318 | USD 0.636 / USD 1.908 |
| IBM | 2 | Granite 4.0 H Small USD 0.0636 / USD 0.265 | USD 0.3498 / USD 0.4505 |
| OpenAI | 1 | OpenAI GPT OSS 120B USD 0.159 / USD 0.636 | — |

## IBM watsonx.ai price list

USD per 1M tokens: 9 prices for 9 models, one row per model with any other service tiers stacked under the standard price. Sources checked 2026-11-15 unless a row says otherwise. "vs maker" compares a standard price with the model maker's own API on the 3:1 blend; a minus means IBM watsonx.ai is cheaper. IBM watsonx.ai is the maker of 2 models here, so there is little maker price to compare with.

IBM watsonx.ai API prices by model
| Model | Tier | Input | Cached | Output | Context | vs maker | Source |
| --- | --- | --- | --- | --- | --- | --- | --- |
| [Granite 4.0 H Small](https://broadrouter.com/llm-api-pricing/granite-4.0-h-small/) | standard | USD 0.0636 | — | USD 0.265 | not stated | own model | [source](https://www.ibm.com/products/watsonx-ai/pricing) |
| [Granite 8B Code Instruct](https://broadrouter.com/llm-api-pricing/granite-8b-code-instruct/) | standard | USD 0.636 | — | USD 0.636 | not stated | own model | [source](https://www.ibm.com/products/watsonx-ai/pricing) |
| [Llama 3.2 11B Vision Instruct](https://broadrouter.com/llm-api-pricing/llama-3.2-11b-vision-instruct/) | standard | USD 0.371 | — | USD 0.371 | not stated | — | [source](https://www.ibm.com/products/watsonx-ai/pricing) |
| [Llama 3.3 70B](https://broadrouter.com/llm-api-pricing/llama-3.3-70b-instruct/) | standard | USD 0.7526 | — | USD 0.7526 | not stated | — | [source](https://www.ibm.com/products/watsonx-ai/pricing) |
| [Llama 4 Maverick 17B 128E Instruct FP8](https://broadrouter.com/llm-api-pricing/llama-4-maverick/) | standard | USD 0.371 | — | USD 1.484 | not stated | — | [source](https://www.ibm.com/products/watsonx-ai/pricing) |
| [Mistral Large 3](https://broadrouter.com/llm-api-pricing/mistral-large-2512/) | standard | USD 0.636 | — | USD 1.908 | not stated | — | [source](https://www.ibm.com/products/watsonx-ai/pricing) |
| [Mistral Medium 3](https://broadrouter.com/llm-api-pricing/mistral-medium-3/) | standard | USD 3.18 | — | USD 9.50 | not stated | — | [source](https://www.ibm.com/products/watsonx-ai/pricing) |
| [Mistral Small 3.1 24B Instruct](https://broadrouter.com/llm-api-pricing/mistral-small-3.1-24b-instruct/) | standard | USD 0.106 | — | USD 0.318 | not stated | — | [source](https://www.ibm.com/products/watsonx-ai/pricing) |
| [OpenAI GPT OSS 120B](https://broadrouter.com/llm-api-pricing/gpt-oss-120b/) | standard | USD 0.159 | — | USD 0.636 | not stated | — | [source](https://www.ibm.com/products/watsonx-ai/pricing) |

-   **Cached "—"** means IBM watsonx.ai's page shows no cache price for that row; budget cached tokens at the input price.
-   **Context "not stated"**: IBM watsonx.ai's pricing page does not give context windows.

[Report a wrong price or number](mailto:info@broadnet.ai?subject=Wrong%20price%20or%20number%20on%20IBM%20watsonx.ai%20API%20pricing&body=Page%3A%20https%3A%2F%2Fbroadrouter.com%2Fllm-api-pricing%2Fproviders%2Fibm-watsonx%2F%0A%0AWhat%20looks%20wrong%20(model%2C%20figure%20shown%2C%20correct%20figure%20and%20a%20source%20link%20if%20you%20have%20one)%3A%0A) (email to info@broadnet.ai, with this page's link filled in).

## What service tiers does IBM watsonx.ai publish?

Standard prices only: IBM watsonx.ai publishes no per-model batch, flex, priority or fast price on the pages we read.

Cached input: no cache-read price is published, so repeated prompts bill at the full input price.

**What we recorded about IBM watsonx.ai's pricing page:** Pay-as-you-go per-1M-token prices from ibm.com/products/watsonx-ai/pricing (US page). IBM: "Prices shown are indicative, may vary by country, exclude any applicable taxes and duties". Models marked "Not available" for pay-as-you-go (hourly deploy-on-demand only) are excluded, as are Granite Guardian / Llama Guard classifiers, speech, vision-only and embedding models. Footnote: "Mistral commercial models have a GPU hosting fee and a model access fee."

## BroadNet Inference and IBM watsonx.ai

BroadRouter is BroadNet's own site, so BroadNet is never ranked on this page. BroadNet Inference serves none of the models IBM watsonx.ai prices. [BroadNet Inference pricing](https://broadrouter.com/pricing/).

## More on IBM watsonx.ai

-   [IBM watsonx.ai's own pricing page](https://www.ibm.com/products/watsonx-ai/pricing) (the source; checked 2026-11-15)
-   [All 50 providers compared](https://broadrouter.com/llm-api-pricing/providers/) · [Cheapest LLM APIs right now](https://broadrouter.com/llm-api-pricing/cheapest/) · [Download every price (CSV, JSON)](https://broadrouter.com/data/llm-api-provider-prices/)

## Questions

### How much does the IBM watsonx.ai API cost per token?

IBM watsonx.ai's standard prices run from USD 0.0636 input and USD 0.265 output per 1M tokens (Granite 4.0 H Small) to USD 3.18 / USD 9.50 (Mistral Medium 3). The median across 9 standard prices is USD 0.371 input and USD 0.636 output per 1M, which is USD 0.000000371 per input token. Input, cached input and output are billed separately.

### What is the cheapest model on IBM watsonx.ai?

Granite 4.0 H Small, at USD 0.0636 input and USD 0.265 output per 1M tokens (USD 0.114 on a 3:1 input-to-output blend). The cheapest from other makers: Llama 3.2 11B Vision Instruct (Meta) at USD 0.371 / USD 0.371; Mistral Small 3.1 24B Instruct (Mistral AI) at USD 0.106 / USD 0.318; OpenAI GPT OSS 120B (OpenAI) at USD 0.159 / USD 0.636.

### Is IBM watsonx.ai cheaper than other providers?

It depends on the model. Of 7 models that other providers also price, IBM watsonx.ai is the cheapest standard price for none; another provider is cheaper for 7; and it is the dearest for 3. Its median rank is 3 of a median 4 providers, on a 3:1 input-to-output blend.

### Does IBM watsonx.ai have batch, cached or priority pricing?

Cached input: no cached price is published on its pricing page, so budget repeated prompts at the input price. No batch, flex or priority prices are listed per model on the pages we read.

### How much does a chatbot cost per month on IBM watsonx.ai?

For 2,000 requests a day, 1,500 input tokens (60% cached), 300 output, over 30 days: USD 10.49 on Granite 4.0 H Small, USD 60.10 on Llama 4 Maverick 17B 128E Instruct FP8 (the middle of IBM watsonx.ai's list) and USD 457.20 on Mistral Medium 3. [Try your own numbers](https://broadrouter.com/tools/llm-api-cost-calculator/).

Try it on your real workload. New accounts start with $5 of credit for Gemma and transcription. DeepSeek, GLM and Qwen3.8 need a top-up.

[Get started](https://inference.broadnet.ai/account/?utm_source=broadrouter&utm_medium=referral&utm_campaign=llm-api-pricing-providers-ibm-watsonx&utm_content=page-cta) ·

**Sources**

-   [IBM watsonx.ai: pricing page (checked 2026-11-15)](https://www.ibm.com/products/watsonx-ai/pricing)
