ibm-granite

IBM: Granite 4.2 8B

IBM's Granite 4.2 8B is a language model that processes text as input and produces output based on its reasoning capabilities. It can handle input contexts up to 131072 tokens and generate up to 117964 completion tokens. The model supports the use of tools and has reasoning capabilities, but it does not provide structured output. For users who prioritize strong reasoning capabilities and are willing to invest in a premium model, IBM's Granite 4.2 8B may be worth shortlisting. Its pricing is $0.1 per million input tokens and $0.15 per million output tokens, which may be competitive for those with large-scale processing needs. Additionally, the model has demonstrated a blended benchmark score of 27.1 across three independent benchmarks, suggesting its performance in various contexts.

Quality Score
91/100
price + capability + benchmarks
Input Price
$0.10
per 1M tokens · checked 2026-09-04
Output Price
$0.15
per 1M tokens · checked 2026-09-04
Context Window
131,072
tokens

Benchmark results

Independent, published benchmarks. Blended score 27.1 across 3 benchmarks, last refreshed 2026-09-04. How scoring works →

BenchmarkMeasuresScore
AI Index broad capability composite 32.4
AI Index Coding software engineering tasks 36.9
AI Index Agentic multi-step tool-using tasks 15.1
Model ID
ibm-granite/granite-4.2-8b
Vendor
ibm-granite
Released
August 2026
Tokenizer
Other
Input Modalities
text
Output Modalities
text
Max Output
117,964 tokens
Tool Calling
✓ supported
Structured Output
✓ supported
Reasoning Mode
✓ supported
Vision
text only
Audio
no
Moderated
no

What it costs in practice

Computed from the current $0.10/M input and $0.15/M output rates. Run your own numbers →

JobTokensCost
Summarize a 50-page report 30k in / 1.5k out under $0.01
Classify 1,000 customer emails 500k in / 50k out $0.06
A month of a busy support chatbot 5M in / 2M out $0.80

Similar models

Quick answers

How much does IBM: Granite 4.2 8B cost?
$0.10 per million input tokens and $0.15 per million output tokens.
What is IBM: Granite 4.2 8B's context window?
131,072 tokens, roughly 196 pages of text in a single request.
Does IBM: Granite 4.2 8B support tool calling?
It supports tool calling, structured output, a reasoning mode.
Can IBM: Granite 4.2 8B process images?
No, it is text-only on the input side.