Verdict
Built from the measurements below
- Accuracy is too close to call: Ministral 3 8B 2512 scored 42.5% and Nemotron 3.5 Super VL 40.0% on the same 80 questions, a gap within what chance alone could produce.
- Ministral 3 8B 2512 is 4.7× cheaper per question as billed ($0.00027 vs $0.0013), reasoning tokens included.
- Nemotron 3.5 Super VL generates text 5.8× faster: 715 tokens per second against 124.
- Ministral 3 8B 2512 starts answering sooner: 0.3 s to the first token, against 0.6 s for Nemotron 3.5 Super VL.
- Nemotron 3.5 Super VL accepts more context: 1M tokens against 262k.
Which should you choose?
Choose Ministral 3 8B 2512 for the lower cost per correct answer ($0.00063 against $0.0032), a quicker first token and shorter outputs (about 1,496 tokens per question against 2,771). Choose Nemotron 3.5 Super VL for faster streaming, a 1M-token context window and more hosts to choose from on OpenRouter (9 against 1). The widest subject gap is in medicine: Ministral 3 8B 2512 answered 5 of 11 correctly, Nemotron 3.5 Super VL 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 = 0.32). We call a winner only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.