Mistral Large 4 vs Mistral Medium 3.5: accuracy, speed and price

• Mistral Large 4 is more accurate: 50.0% against 28.8% on the same 80 questions, a 21.3-point gap that is statistically significant.
• Mistral Large 4 is 2.2× cheaper per question as billed ($0.011 vs $0.025), reasoning tokens included.
• Mistral Medium 3.5 generates text 1.7× faster: 145 tokens per second against 83.
• Mistral Medium 3.5 starts answering sooner: 0.5 s to the first token, against 1.1 s for Mistral Large 4.
• Mistral Large 4 accepts more context: 1M tokens against 262k.

Which should you choose?

Choose Mistral Large 4 for the most accurate answers of the two, the lower cost per correct answer ($0.023 against $0.087) and a 1M-token context window. Choose Mistral Medium 3.5 for faster streaming and a quicker first token. The widest subject gap is in science: Mistral Large 4 answered 20 of 30 correctly, Mistral Medium 3.5 5 of 30. With so few questions per subject, treat this as a lead to test on your own prompts, not a result.

Accuracy gap +21.3 points (z = 2.75). We call a winner only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.