What I most want to see it compared to is Gemma 4 12B in the 4-bit QAT version. It's barely bigger than this at just under 7GB, so it also runs on just about any modern device and is remarkably smart for its size. It's an excellent tool user, crazy good vision for its size. I'm still trying to wrap my head around how much is lost with each step down in resolution, but the QAT versions from Google seem to prove the answer is "very little" at four bits.
4bits is a cutoff point for many model families, but also depends on what parts you quant to 4bits vs alternatives (weights, weight+activation, kv cache).
I'm currently working towards an updated version (not an og author), curious if others are aware of similar surveys, as I have yet to do a real lit search.
I need help understanding this.
I understood that the magic here is the quantization that allows it to use from 50G to 4G and their process retain most of the intelligence within Pareto limits of gain. And then they proceed to compare with other quantized models as in the level of intelligence per size. It gets to my attention though that the performance in tool calling is mostly affected which is a problem for other small models.
How does this model compare to a recent 4G model? How do we know it retained intelligence from the parent rather then being fine tuned for the benchmarks?
I am not shtng on them or anything. I'd rather find it amazing, BUT given my limited knowledge, I feel the results miss fair comparison plots and the ones might be misleading. Buy I also reckon it might be me the problem.
Anyone care to explain this poor silly fellow some of those points?
from what I understand prismml isn’t doing a quant like normal models where you take a model trained at fp16 and then chop off some bits to reduce vram, but rather they’re training the model natively with 1 bit weights. It’s explained more in the article. They’re also doing some other tricks like a fp16 weight per block of 128 1bit weights to get some more data out of 1 bit weights
Notably, PrismML CEO Babak Hassibi told CNBC this, so it’s either (1) bullshit, or (2) he just ended any chance of a relationship by leaking news of the talks.
Apple would punish him severely unless they cleared it in advance, it might be to their advantage for some reason (negotiating with Google for Gemma rights? idk).
Awesome! I've been waiting for them to start scaling ternary models for over a year[1]. Excited to try it out, typical Qwen 27B is too heavy for me to run on my local hardware at reasonable speeds.
Quite weird that heavy quantization method on a dense model gives better results than slightly quantized MoE models like 35B-A3B from Google.
At this point all the different quantization and 'compression' (look at MPO applied to LLMs...) techniques start feeling a bit like snake oil. It's just gut feeling - or scores on benchmarks models are optimized for - what ends up deciding whether a technique is good enough or not.
I've tried a couple in LM Studio - the GGUF one and the MLX one - but neither worked there. Anyone else get them to work? Might be that LM Studio needs to upgrade their llama.cpp or MLX engines first.
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.
quanting kv cache hurts attention / recall, and long-form tasks by proxy. Model families and sizes have different tolerances to quant ting different parts of the model, same for intended tasks.
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?
If you read to the bottom of the page, it says they're funded by a few people, and one of them is Samsung. I'm betting Samsung wants to be able to ship a capable AI system on a future model of their phone so they can compete with Apple.
Agreed, and the prevailing wisdom now seems to be that unless you can release a truly frontier model, you might as well release yours as open source to undercut your competition.
It appears they are using Q2_0 in llama.cpp, which is 2 bits per weight + 1 float16 scale per group of 64 weights. This is inefficient in two ways: one bit pattern is wasted on each weight, since ternary weights only use {-1,0,1} and Q2_0 allows {-1,0,1,2}; and their group size is 128 weights, so the scale will be stored twice in two groups of 64 instead of stored only once in one group of 128.
Their fork corrects the second inefficiency by using a group size of 128, but still uses 2-bit weights AFAICT.
It's possible to pack 5 trits into a byte, but the unpacking is not very efficient. Another recent idea is to add the constraint that exactly one weight in each group of four be zero, which gives exactly 32 possible states, so it fits in 5 bits.
You can beat the efficiency of 5 trits in 8 bits (1.6) with as few as 17 trits in 27 bits (~1.588), but once you account for rounding up to a whole number of bytes for practical reasons, then beating the efficiency requires going to at least 111 trits in 176 bits (~1.586), or perhaps more practically for fast unpacking, 161 trits in 256 bits (~1.59).
At that level, even if you have, say, 27B trits, the more efficient encodings would save something like 38-45MB (theoretical limit ~48MB), likely at the cost of some slowdown.
It’s still a bit with only two possible values. But they add a scaling factor to a group of them (128 for example) which when you factor in, results in a fractional number of bits per parameter.
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.
After using a highly capable 2-bit quant as my daily driver for months now, I get pretty excited about releases like this. After a few days for the kinks to be worked out, I’ll be excited to try it.
I find these style of models are great, but fail hard, and fail randomly. I'd be hesitant to use it for a daily driver, but I'm using dual 3060s, so it's not like I'm quantizing a frontier model here.
How do you find the overall experience? And do you have any special sauce or recommendations for going this route?
I’m using DeepSeek V4 Flash on 128gb mbp - it’s a bit different using a 200b+ param model. It’s MoE so performance is acceptable. It will still malform a tool call every now and then, but the capabilities are so far ahead anything else that the majority of the time it works really well and solves really complex problems.
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)
Looks like they quant'd too hard at 4 bits, can't imagine the ternary being any good based on this. I'm also not sure what is up with the gsm8k, their benchmarks show something different, but they are using another eval tool. I'll have to add it to my setup. Also why I'm building a setup instead of taking model devs word for benchmarks. (https://github.com/modelscope/evalscope)
Code if you'd like to reproduce or try other test sets: https://github.com/verdverm/quantr (lightly tuned to a single oem spark, probably possible in 32-48G)
Good paper to understand the effects of quant regimes across model families and tasks: https://arxiv.org/abs/2402.18158 (Evaluating Quantized Large Language Models - 2024 ICML)
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.
Of course not, personally almost all of my code these days is generated.
The LLM style of writing is just very distracting to read. “It unlocks X”, “Y changes the equation”, and why is there always something shifting? Makes my eyes glaze over in an otherwise interesting post.
Yep, that’s the question. I asked just that when Bonsai’s first models got released. Super interesting if we can push the parameter count over 100B with 1.125 bit quantization and still keep pretty good performance versus 16-bit 100B models. That’s a definite sweet spot.
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.
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.
Oh, I don't actually know the difference if you want to explain it
The title says it's 27B grade running on a phone and what I was comparing it to in my mind was a model that runs at 35B grade that could presumably run on a phone "better"?
edit: I asked AI for the difference and understand a little better, thanks for the heads up to learn the difference between models... I think the thing was, although ornith was created for a specific agentic purpose, it was still outperforming a previous generalist model I had running locally (so in my mind I thought it was still a better local model) - I'd like to try bonsai out if I can figure out how to run it lol
Orinth was not impressive in my vibes testing, I just completed my first grid analysis with real evals on qwen 27b. I can now scale that grid analysis and intend to include the qwen 9b ftunes I've seen going around. They were actually a main motivation because so many claim this or that one is better, but very little in the way of evals
27B is way more than you need for a phone. Doesn't matter how much you try to compress it, it's the wrong application of the wrong tool. There are already useful tiny models that fit on phones and do basic things really well. Dumb down a big model too much and it becomes worse than a small fine-tuned model.
The article is about running it on a phone though, and shows an app with their branding running this in text mode on a phone. I'm asking where can I find this app to try what is being demonstrated in this article & video? Appstore only has an image gen app by them and other MLX apps I've tried don't seem to support this model
Why make this comment without having tried it first? It very clearly is not useless and performs a lot better than one might expect. I am currently waiting to do more benchmarks of it in comparison to the full weight model, but it seems promising/better than Mistral Nemo at a lower file size.
I think what OP means is that the "minimum viable product" for a daily use LLM is probably somewhere around e.g. GPT 4o's level of intelligence (YMMV). Below a certain threshold, you are better off using specialized machine learning models rather than general purpose LLMs. It's very difficult to get that level of intelligence fully local on a mobile device without streaming to the cloud.
I do think this would be interesting if they made these easy to finetune, as I do think this level of intelligence is likely sufficient for many applications and could be extremely cheap to run.
Good evaluation from 2024 https://arxiv.org/pdf/2402.18158
I'm currently working towards an updated version (not an og author), curious if others are aware of similar surveys, as I have yet to do a real lit search.
How does this model compare to a recent 4G model? How do we know it retained intelligence from the parent rather then being fine tuned for the benchmarks?
I am not shtng on them or anything. I'd rather find it amazing, BUT given my limited knowledge, I feel the results miss fair comparison plots and the ones might be misleading. Buy I also reckon it might be me the problem. Anyone care to explain this poor silly fellow some of those points?
https://www.theregister.com/on-prem/2000/08/02/jobs-snubs-at...
apple’s secrecy agenda has been defeated to an extent by the practicalities of ubiquitous technology?
[1] https://jackson.dev/post/dont-sleep-on-bitnet/
At this point all the different quantization and 'compression' (look at MPO applied to LLMs...) techniques start feeling a bit like snake oil. It's just gut feeling - or scores on benchmarks models are optimized for - what ends up deciding whether a technique is good enough or not.
I've tried a couple in LM Studio - the GGUF one and the MLX one - but neither worked there. Anyone else get them to work? Might be that LM Studio needs to upgrade their llama.cpp or MLX engines first.
Details are here -> https://github.com/PrismML-Eng/Bonsai-demo/blob/main/README....
You can also join Discord to communicate with us directly http://discord.gg/prismml
I wish KV-cache memory usage and related optimizations were discussed more clearly in new model announcements and demos.
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.
is it a float? if so, how many bits is the float?
I've never heard of a bit ever having more than two possible values
Their fork corrects the second inefficiency by using a group size of 128, but still uses 2-bit weights AFAICT.
It's possible to pack 5 trits into a byte, but the unpacking is not very efficient. Another recent idea is to add the constraint that exactly one weight in each group of four be zero, which gives exactly 32 possible states, so it fits in 5 bits.
It's not represented by a "bit", binary digit with value of 0 or 1; but with a "trit", ternary digit with value of {−1, 0, +1}.
e.g. 5 trits (243 states) into a byte gives 1.6 bits per trit: https://compilade.net/blog/ternary-packing
You can beat the efficiency of 5 trits in 8 bits (1.6) with as few as 17 trits in 27 bits (~1.588), but once you account for rounding up to a whole number of bytes for practical reasons, then beating the efficiency requires going to at least 111 trits in 176 bits (~1.586), or perhaps more practically for fast unpacking, 161 trits in 256 bits (~1.59).
At that level, even if you have, say, 27B trits, the more efficient encodings would save something like 38-45MB (theoretical limit ~48MB), likely at the cost of some slowdown.
The way they do it is packing like the other comment says.
Each byte represents 5 trinary values instead of 8 binary, and there is a little bit of waste.
https://github.com/Liquid4All/antidoom
Do they have plans to bring even bigger models down to ~16GB VRAM so that more consumer hardware might be useful?
Sounds like the model is not following a proper probabilistic choice here, so maybe more a programming error than a model training error.
When I saw 27b on a phone, I thought not fitting, big phone, or aggressive quant. NVFP4 still takes 27G before KV cache.
I find these style of models are great, but fail hard, and fail randomly. I'd be hesitant to use it for a daily driver, but I'm using dual 3060s, so it's not like I'm quantizing a frontier model here.
How do you find the overall experience? And do you have any special sauce or recommendations for going this route?
The 2 bit quants are really good. I have a lot of memory so I can squeeze it all in at ~80gb.
Code if you'd like to reproduce or try other test sets: https://github.com/verdverm/quantr (lightly tuned to a single oem spark, probably possible in 32-48G)
Good paper to understand the effects of quant regimes across model families and tasks: https://arxiv.org/abs/2402.18158 (Evaluating Quantized Large Language Models - 2024 ICML)
The LLM style of writing is just very distracting to read. “It unlocks X”, “Y changes the equation”, and why is there always something shifting? Makes my eyes glaze over in an otherwise interesting post.
> 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
The title says it's 27B grade running on a phone and what I was comparing it to in my mind was a model that runs at 35B grade that could presumably run on a phone "better"?
edit: I asked AI for the difference and understand a little better, thanks for the heads up to learn the difference between models... I think the thing was, although ornith was created for a specific agentic purpose, it was still outperforming a previous generalist model I had running locally (so in my mind I thought it was still a better local model) - I'd like to try bonsai out if I can figure out how to run it lol
huggingface.co/Tivaphraen/Geryon-9B-v1
I can just see their image tool on the app store
Available on HuggingFace: https://huggingface.co/collections/prism-ml/bonsai-27b
The article is about running it on a phone though, and shows an app with their branding running this in text mode on a phone. I'm asking where can I find this app to try what is being demonstrated in this article & video? Appstore only has an image gen app by them and other MLX apps I've tried don't seem to support this model