> > anon asked for one more state
 > > we added zero
 > > +600 MB
 > > +5 benchmark points
 > > 75.5 avg at 1.75 GB
 > > still ~1/9 the size of Qwen3 8B
 > > shout out brahmagupta
 > > zero mattered

Turns out adding 0 helps :) Today we’re introducing Ternary Bonsai 🌳, a family of end-to-end 1.58-bit language models in 8B, 4B, and 1.7B sizes. Ternary Bonsai 8B is within 5% of Qwen 3 8B at 9x lower memory. Still tiny. Noticeably smarter

Awesome release, congrats to the. PrismML team! It comes with day-0 support on MLX thanks to some of the work we did with bitnet-1.58 kernels a year ago. huggingface.co/collections/pr…

People are the most valuable resource in action.

Ternary Bonsai: state-of-the-art intelligence at 1.58 bits. The models are so small they can even run locally in your browser on WebGPU! ⚡️ Here's the 8B version (just ~2GB in size) running at 60 tokens per second on my M4 Max. Try the demo out yourself! 👇

New models from PrismML. Incredible intelligence density. The new models outperforms other models in comparable parameter classes. Available on iPhone in Locally AI - Local AI Chat.

Thanks merve and the Hugging Face team!

Check out how you can use Ternary Bonsai 8B 🌳 for tool calling in your everyday life—an impressive demo on an amazing platform by AnythingLLM and Tim Carambat!

The models are getting smaller. Great for OpenClaws and Hermes. Gotta heat them up! Yesterday someone told me "phones are three to five years away." Oh, really?

One of the things I tried researching but found really hard. 1.58bpw is insane 10x smaller than original, I hope they push it to much larger models

Ternary Bonsai 8B is within 5% of Qwen 3 8B at 9x lower memory! Congratulations PrismML on yet another exciting release! cc Khosla Ventures

Interesting work here 👇

Ternary is actually surprisingly powerful. Validated by bitnet and now again here. In the new model training research/experimentation I've been working on, ternary weights (in some places) actually beats bf16 (by a not-insignificant amount), at least up to the 7b scale (and

PrismML Ran Ternary-Bonsai 8B on my iPhone through OnDevice LLM. Surprisingly fast.

Tiny local models like Bonsai are going to change things. For the last three years, the default way most people used AI was simple: frontier models lived in data centres, you reached them through an API, and anything local felt like a toy. That will probably stop being true in

Pico Local AI Server 1.4.21 is now available on the Mac App Store. This release adds support for Ternary Bonsai, a lightning-fast model that outperforms many much larger models

Huge shoutout to PrismML This shouldn’t be possible: a tiny model punching way above its weight. The largest version is just 1.14 GB, which means it’s small enough for a phone. Fast on a phone (spoiler: Pico for iOS is coming soon!). Insanely fast on a MacBook Pro M1 Max.

Training gets the headlines. Inference gets the bill. As agents move from novelty to default workload, the hard problem isn't the model anymore. It's every millisecond and every watt between a prompt and the next token. A coding agent running for six hours straight is a very