90 points by xenova 1 hour ago | 14 comments
erwan577 1 minute ago
The KV-cache memory usage also seems remarkably frugal, even at the full context length. That could make this model particularly useful in multi-agent coding workflows.

I wish KV-cache memory usage and related optimizations were discussed more clearly in new model announcements and demos.

sigbottle 23 minutes ago
What's the hiring space and business strategy around all of these smaller AI labs? Its really cool that people like these guys get paid to optimize models and give them out for free (open source). Do a lot of these labs have forward deployed engineers doing integrations with customers who want local models? Is there a general shift towards the local model crowd?
kristianp 7 minutes ago
Apparently Apple is "in talks" with the PrismML: https://www.cnbc.com/2026/07/14/apple-prismml-ai-compression...
liuliu 1 hour ago
The problem, of course, is if you run the UD_Q2 variant (Unsloth) which does only post-training, the number is pretty close to 1-bit model here and the 5% drop in tool-call is significant than it suggests in real-life use cases.
liuliu 1 hour ago
You also need to pay close attention to BFCLv3 multi-turn result, that helps you to get a sense how frequently these quants will be in a doom loop.
syntaxing 39 minutes ago
For those curious about their demo, I’m pretty sure it’s using Locally AI (iOS only) that lmstudio acquired/aquihired a couple months ago.
simonw 57 minutes ago
The models themselves are showing up on Hugging Face here: https://huggingface.co/prism-ml/models
luckystarr 6 minutes ago
Tried it on Android and got "!!!!!!!!!!!!!" for answers.
syntaxing 25 minutes ago
I don’t know if the llama cpp implementation is wonky (and only supports the binary version) but it’s a lot slower than 35B-A3B @ Q4_KM + MTP with CPU offloading.
pulse7 8 minutes ago
Most probably not optimized yet for this model...
thomasjb 19 minutes ago
I've been watching and waiting for this, interested to see how smart it is, as it fits with my interest of getting the smartest possible model running in 10GB of VRAM (RTX3060 that has to drive 2 monitors and run an llm)
alvatech 1 hour ago
TIL that 1 bit models are actually 1.58 bit with three values +1, 0 and -1
NitpickLawyer 56 minutes ago
There's two variants of this (or, as the joke goes, for very big values of bit):

Ternary Bonsai 27B uses ternary {−1, 0, +1} weights with FP16 group-wise scaling, giving a true 1.71 effective bits per weight.

1-bit Bonsai 27B uses binary {−1, +1} weights with the same group-wise scaling, giving 1.125 effective bits per weight.

bensyverson 1 hour ago
Yeah, it's an unfortunate convention from the very first "1 bit" model. But to be clear, Bonsai comes in both ternary and actual 1-bit variants.
erelong 39 minutes ago
I was trying Ornith 9B locally (it's up on Ollama) which claims:

> Ornith-1.0-9B, which can be easily deployed on edge devices, matches or exceeds the performance of much larger models such as Gemma 4-31B and Qwen 3.6 35B.

https://deep-reinforce.com/ornith_1_0.html

Only tried it so much so far; it did a little better than Qwen 9B

janalsncm 24 minutes ago
Is that a 1-bit LLM? I don’t understand the connection with this article.
liuliu 37 minutes ago
Note that 3.5 9B cannot do thinking (while 3.6 27B can, pretty effectively, quite verbosely).
xyzsparetimexyz 49 minutes ago
That's awesome. What's the largest model that could fit onto a single 16gb gpu at 1.125 effects bits per weight?
Catloafdev 14 minutes ago
Doing some naive math, the F16 filesize is ~53.8gb, the 1-bit version is ~3.8gb, about 7% of the original size. The F16 size is roughly 2x param count, so that gives a rough ballpark of ~110B.
Havoc 58 minutes ago
This must be some sort of unpublished app?

I can just see their image tool on the app store

Catloafdev 11 minutes ago
It's a LLM model, not a phone app.

Available on HuggingFace: https://huggingface.co/collections/prism-ml/bonsai-27b

ai_fry_ur_brain 1 hour ago
[dead]