AI Benchmark Model

Granite 4.2 8B

Benchmark scores, pricing, speed, and model comparisons for Granite 4.2 8B.

Key scores

Intelligence19.6Index score
Coding22.4Index score
Speed105.0Tokens/sec
Blended Cost$0.11Per 1M tokens

Granite 4.2 8B is an artificial intelligence model developed by IBM and released on August 25, 2026. The model features an intelligence index of 19.6 and a coding index of 22.4. Performance metrics indicate an output speed of 104.966 tokens per second with a time to first token of 0.236 seconds. Pricing for the model is set at $0.06 per million tokens for input and $0.25 per million tokens for output, resulting in a blended rate of $0.11 per million tokens. Benchmark results for the model include a GPQA score of 0.631, an HLE score of 0.097, a SciCode score of 0.304, and an LCR score of 0.433.

Benchmark results

4 tracked benchmark rows for Granite 4.2 8B.

GPQA Reasoning
63.1%
LCR Other
43.3%
SciCode Other
30.4%
HLE Other
9.7%

Comments (0)

No comments yet

Be the first to share your thoughts!