SOTA Language Models Under 14B?

<!-- SC_OFF --><div class="md"><p>Hi. What do you think is the smallest Open model, as of Spring 2026, that can match/beat OpenAI GPT 4o Mini in the broadest sense possible?</p>

<!-- SC_OFF --><div class="md"><p>I need SOTA ai at like 2k TPS with tiny latency so that I can get time to first answer token under 3 seconds for real time replies with full COT

<!-- SC_OFF --><div class="md"><p>Long time Claude Opus user, but after the recent session limit changes by Anthropic, I am seriously considering trying Chinese models for coding.

<!-- SC_OFF --><div class="md"><p>Tomorrow, I am getting delivery of a 13&quot; M5 macbook air with 32gb RAM and 1T SSD. Currently, I have pro subscriptions to Gemini, Perplexity

<!-- SC_OFF --><div class="md"><p>Hello!</p> <p>I’m working on RAG system on arabic documents any idea on the best embedding model out there?</p> </div><!-- SC_ON --> &#32;

<!-- SC_OFF --><div class="md"><p>Hello. I'm ordering a small piece of software for personal needs (like a virtual keyboard or an expression recognition action app). I asked

<!-- SC_OFF --><div class="md"><p>LLMs keep improving.<br /> Small models are around 3 years behind frontier models.<br /> Do you think we’ll have a model as good as today’s Fable

<!-- SC_OFF --><div class="md"><p>Not sure if anyone out there is working on this, don't see any on huggingface to try out. It might be too new at the moment, but it would be cool