Ministral 3 14B Reasoning

This is the Reasoning post-trainingAny training that happens after pretraining to make a base model useful for real tasks. Includes instruction tuning, chat tuning, and alignment work. Post-training is dramatically cheaper than pretraining — thousands to low millions rather than tens of millions. Most of what distinguishes GPT-4 from Llama 3.1 as a product, rather than as a base capability, is post-training. variant of Ministral 3 14B. The headline benchmark result — 85% on AIME 2025, versus Qwen3-14B's 73.7% — makes this one of the strongest small reasoning models available at any price. The tradeoff is output length: Reasoning variants produce extended chain-of-thought output before their final answer, which means higher tokenThe basic unit of text a model reads and writes. Tokens are roughly three-quarters of a word in English — so 100 tokens is about 75 words. Models don't see letters or words directly; they see tokens. Pricing is almost always quoted per million tokens, and context windows are measured in tokens rather than words. consumption per query. For fast assistant-style interactions, use Ministral 3 14B Instruct (full catalog entry) instead; for math, logic, and complex multi-step reasoning tasks where accuracy matters more than speed, this is the right variant.