Calculate the cost of using granite-3-8b-instruct from IBM watsonx for your AI applications
Pricing data last updated:
Mode: Chat
Max: 8,192 tokens
Max: 1,024 tokens
Built for enterprise-grade reliability, governance, and scale. Deploy in seconds.
granite-3-8b-instruct is a chat model from IBM watsonx, one of 28 chat models they offer. It is priced at $0.20 per 1M input tokens and $0.20 per 1M output tokens, ranking 597 out of 2172 chat models by cost and cheaper than 68% of models in this category. granite-3-8b-instruct supports function calling, prompt caching, structured output. It accepts up to 8K input tokens.
Note: Use the interactive calculator above to estimate costs for your specific usage patterns.
Use the maximum token limits shown above to understand the model's capacity. This model can handle up to 8,192 input tokens. The maximum output length is 1,024 tokens.
At $0.20 per 1M input tokens and $0.20 per 1M output tokens, granite-3-8b-instruct ranks 597 out of 2172 chat models by input cost. It is more affordable compared to the median of $0.55 for chat models, and is cheaper than 68% of models in this category.
granite-3-8b-instruct is one of 28 IBM watsonx chat models, with support for function calling, prompt caching, structured output.
| Model | Provider | Input / 1M tokens | Output / 1M tokens | vs granite-3-8b-instruct |
|---|---|---|---|---|
| gpt-5.4-nano | Azure | $0.20 | $1.25 | 0% |
| gpt-5.4-nano-2026-03-17 | Azure | $0.20 | $1.25 | 0% |
| gpt-4o-mini-2024-07-18 | Azure | $0.17 | $0.66 | -17% |
Similar chat models from other providers
No. granite-3-8b-instruct costs $0.20 per 1M input tokens while gpt-5.4-nano costs $0.20 per 1M input tokens, making gpt-5.4-nano 0% more affordable for input. However, granite-3-8b-instruct may offer different capabilities or performance characteristics that justify the price difference.
granite-3-8b-instruct input pricing is $0.20 per 1M tokens, which is 64% below the median of $0.55 for chat models. It ranks 597 out of 2172 chat models by input cost, making it cheaper than 68% of models in this category. For output, it costs $0.20 per 1M tokens compared to the median of $1.48.
Among IBM watsonx's 28 chat models, granite-3-8b-instruct ranks 9 by input cost.
The most comparable chat models to granite-3-8b-instruct are: gpt-5.4-nano from Azure ($0.20/1M input tokens); gpt-5.4-nano-2026-03-17 from Azure ($0.20/1M input tokens); gpt-4o-mini-2024-07-18 from Azure ($0.17/1M input tokens); gpt-4o-mini from Azure ($0.17/1M input tokens). These alternatives were selected based on similar capabilities, pricing, and provider diversity. You can compare any of these models in detail using the Bifrost Model Library.
granite-3-8b-instruct is priced based on input and output tokens. Use the interactive calculator at the top of this page to estimate costs for your specific workload. Enter your expected input and output tokens volume and the calculator will show the total cost breakdown. For reference, processing 1M input tokens costs $0.20 and generating 1M output tokens costs $0.20.
