AI Rate CardBase model
granite-7b-instruct
from $0.40 at 1 provider
ibm/granite-7b · 7B parameters · 1 provider · Text models ·
Price alerts
Found 1
Simple Full
Ranges in one field: "from-to" - "0.5-3", "-3" (up to), "128k-" (from); a colon works too. Several ticked values of one column are ORed, columns are ANDed.
Price history is collected since Nov 13, 2026.
The GRANITE family - base models: 17
- granite
- granite-3.0-3b-a800m-instruct 1
- granite-3-0-8b-instruct 1
- granite-4.2-3b 4
- granite-4.2-30b 4
- Granite-4.0-H-Micro 6
- granite-4-micro 1
- IBM: Granite 4.0 Micro 1
- Granite 4.1 8B 1
- Granite 4.2 8B 9
- granite-4h-small 1
- granite-7b-instruct 1
- granite-7b-base 1
- granite-7b-lab 1
- granite-8b-code-instruct 2
- granite-34b-code-instruct 1
- granite-embedding-278m-multilingual 1
- granite-guardian-3-8b 1
About the model: cost of your workload, questions, comparisons, citation
Cost of my workload → in the Tools tab: every provider of this model, cheapest first
Frequently asked questions
How much does granite-7b-instruct API cost?
From $0.40 input and $0.80 output per 1M tokens, at featherless.
Which provider offers granite-7b-instruct cheapest?
featherless offers the lowest price, $0.40 per 1M tokens, among 1 provider.
What is the context window of granite-7b-instruct?
4K tokens.
Source: AI Rate Card, https://airatecard.com/en/model/ibm/granite-7b, prices checked 2026-11-14