Granite 4.0 Micro vs Granite 4.2 8B: accuracy, speed and price

• Accuracy is too close to call: Granite 4.0 Micro scored 25.0% and Granite 4.2 8B 36.2% on the same 80 questions, a gap within what chance alone could produce.
• Granite 4.0 Micro is 1.5× cheaper per question as billed ($0.000051 vs $0.000077), reasoning tokens included.
• Granite 4.2 8B generates text 1.6× faster: 60 tokens per second against 38.
• Granite 4.2 8B starts answering sooner: 0.6 s to the first token, against 0.7 s for Granite 4.0 Micro.

Which should you choose?

Choose Granite 4.2 8B for faster streaming, a quicker first token and shorter outputs (about 227 tokens per question against 403). The widest subject gap is in economics: Granite 4.2 8B answered 5 of 9 correctly, Granite 4.0 Micro 2 of 9. With so few questions per subject, treat this as a lead to test on your own prompts, not a result.

Accuracy gap -11.2 points (z = -1.54). We call a winner only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.