Key Info

Carbon-X is a 14B open model that combines DNA and natural language, mid-trained on 300B tokens of interleaved DNA and text, then SFT and RL, released under Apache 2.0.

Highlights

  • Beats every open language model tested on DNA benchmarks and comes within a point of GPT-6.1 Sol at over 1,000x lower inference cost.
  • Uses a new hybrid tokenizer encoding DNA with roughly 3x fewer tokens than BPE.
  • Can act as a DNA specialist subagent: Kimi K3 gains 11.1 points when consulting it.
  • Runs locally on a laptop via llama.cpp, keeping DNA data on local hardware.