Nori LLM claims to be the fastest model, achieving over 1 million tokens per second, but is noted as being confidently inaccurate and unreliable for important information.
MiMo-V2.6-Pro is an open-weight AI model by Xiaomi released in September 2026, scoring 46 on the Artificial Analysis Intelligence Index with strong performance at 125 tokens per second. Priced at $0.43 per 1M input tokens and $0.87 per 1M output tokens, it supports multimodal inputs (text, image, speech, video) with a 1M token context window.
Jev is a decision-only AI model that outputs structured classifications (yes/no, category picks, numeric scores) rather than text, built using RLCD by TypeSafe AI. Testing across 12 use cases showed it is 20–200x faster and 40–400x cheaper than standard LLMs for classification tasks, with optimal results in a two-model pipeline where Jev triages high-volume data before routing to GPT or Claude for reasoning.