50GRAMx Think

Open models on your machines.

Run open models for text, speech, vision and embeddings on the grams you own, through the engines they already use.

How it works

  • Grams run llama.cpp, Ollama and vLLM for text, and speech engines for listening and speaking.
  • An engine counts as healthy only when it answers; a crash-looping engine is not offered.
  • A small or battery gram hands a turn it cannot take to one of your always-on grams instead of dropping it.
Built in
  • 2026-11-07 · epn-daemon ec7a5d46
  • 2026-11-07 · epn-daemon f3275b62
  • 2026-11-10 · epn-daemon 3ba6b84b
Next: Measured before you chooseQuickstartAll Think features
Start with Evo

Evo works on your own gram, with your machines in view, and asks before every change. Take one of these to it, or write your own.

  1. 1. Run a gramOn a machine you have. Evo lives there.
  2. 2. Enter your gramxYour business on the network.
  3. 3. Ask EvoPaste the question; it works through it with you.