Gemma 4 31B vs Mistral Small 4: accuracy, speed and price

• Accuracy is too close to call: Gemma 4 31B scored 30.0% and Mistral Small 4 32.5% on the same 80 questions, a gap within what chance alone could produce.
• Gemma 4 31B is 1.3× cheaper per question as billed ($0.0014 vs $0.0019), reasoning tokens included.
• Mistral Small 4 generates text 2.9× faster: 170 tokens per second against 58.
• Mistral Small 4 starts answering sooner: 0.5 s to the first token, against 0.6 s for Gemma 4 31B.

Which should you choose?

Choose Gemma 4 31B for the lower cost per correct answer ($0.0046 against $0.0057) and more hosts to choose from on OpenRouter (13 against 1). Choose Mistral Small 4 for faster streaming and a quicker first token. The widest subject gap is in medicine: Mistral Small 4 answered 5 of 11 correctly, Gemma 4 31B 2 of 11. With so few questions per subject, treat this as a lead to test on your own prompts, not a result.

Accuracy gap -2.5 points (z = -0.34). We call a winner only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.