This is super interesting, I'm particularly excited for this one as it may allow teams to scale this architecture for VLAs (vision language action models), and having sparser models means more real-time actions on a locally hosted model
Why does this not have (day-one) support for Ollama? The previous model is on there? Is it related to the ongoing refactor work or are people abandoning Ollama for other LLM engines?
I'd normally call that a low-effort, troll comment. But, thinking on it, you may have a great metaphor.
They keep promising great performance out of models whose key ingredient (parameters) they are diluting. Many seem to be in a competition saying they're getting smaller and higher performance at the same time. Then, the homeopathic models don't perform as well as real models when independently tested. Again, spot on.
Wow, this is fucking phenomenal. I fed it a long transcript asking it to create a summary and it executed it extremely well. For an 8B model this is quite impressive.
Many such cases. Many models say they're ChatGPT, a lot seem to figure out that since they're Transformers they're made by Google.
Doesn't really tell you a lot. Perhaps a pretraining / midtraining artifact.
I really love how fast it is! Their press release comparing it on Strix Halo and M5 Max are impressive. It going twice as fast at GPU benchmarks even more so!
I just tested this on a bug fixing benchmark I'm working on.
It did not perform as well as I expected. Qwen2.5-Coder-3B (2 years old) outperformed it by a wide range -> fixing ~50% of bugs whereas this model only fixed ~12%.
Granted, it's not a coder specific model, but given its benchmark performance to Gemma models, and that it's two years newer, and that it's an MoE with 8B total params, I expected it to be more competitive.
I will test it when it's accessible via OpenRouter, but the previous LFM2 model (lfm-2-24b-a2b) didn't do well on my tests, it got only 1/20 questions/tasks right, way below Gemma 31B or Qwen 35b-a3b (those get like 10/20 right)
I personally find any model smaller than something like Qwen 3.6 35B-A3B (8-bit quantization, about 49GB memory usage when loaded into llama.cpp) to be too "stupid" for reliable use.
I would much rather not run the model on my local laptop hardware and offload that to some system sitting under my desk in my home office, accessible via VPN, than take the risk of using an unreliable and flaky tool for the convenience of having it on the same hardware on my lap.
I pay very little attention to 8 billion or whatever (or even much smaller) models these days and I don't feel like I'm missing much.
That's not all that surprising, IMO. From what I understand, LiquidAI is focusing pretty narrowly on building models that operate as the "agentic core" of a larger system.
If I were going to use this model, I'd be looking to use it more as is the primary chat interface of a larger system, and having it orchestrate & delegate tasks to other places via tool calls. It's not quite as exciting on the surface as a local "do it all" model, but it does enable some pretty neat use-cases, IMO.
I'm imagining a local agent that is super low latency, works entirely offline, and capable of queuing up complex tasks for larger/smarter cloud agents which execute them asynchronously.
Look at the accuracy numbers and these things clearly don't know much yet, and I'm not about to hand one my hardest work. But you can see where it's going. As quantization and the MoE stuff keeps getting better, "good enough to just run on my own machine" keeps eating into more of what I'm currently paying a frontier lab for. Once a local model can handle like 80% of what I need, the math stops making sense for the subscription.
At some point we have to be running into some inherent mathematical limits of knowledge compression, right? No way the knowledge benchmarks on these 8B models will keep getting better without overfitting on these benchmarks
Beware the license. They misleadingly state on the blog post "Open-weight — Download, fine-tune, and deploy without restrictions". But if you read their license <https://huggingface.co/LiquidAI/LFM2.5-8B-A1B/blob/main/LICE...> it has significant restrictions for any org with other $10M in revenue.
33 comments
[ 4.5 ms ] story [ 62.2 ms ] threaddemo link for anyone that wants to try this out https://playground.liquid.ai/chat?model=cmppnbgse000004l4bc8...
They keep promising great performance out of models whose key ingredient (parameters) they are diluting. Many seem to be in a competition saying they're getting smaller and higher performance at the same time. Then, the homeopathic models don't perform as well as real models when independently tested. Again, spot on.
Answer: . . . . So, unless you have a compelling reason not to, walk to the car wash.
I recently realized that Qwen3.5:4B is way more capable than I thought a model that size could be.
Combine that with the work Liquid puts into RL and fine tuning, and you get models that perform extremely well on minimal hardware.
Combine that with your own fine tuning, and you get a specialized tool that is fast, private, and doesn’t require internet connection.
[1] https://github.com/Liquid4All/cookbook/tree/main/examples/lo...
It did not perform as well as I expected. Qwen2.5-Coder-3B (2 years old) outperformed it by a wide range -> fixing ~50% of bugs whereas this model only fixed ~12%.
Granted, it's not a coder specific model, but given its benchmark performance to Gemma models, and that it's two years newer, and that it's an MoE with 8B total params, I expected it to be more competitive.
I would much rather not run the model on my local laptop hardware and offload that to some system sitting under my desk in my home office, accessible via VPN, than take the risk of using an unreliable and flaky tool for the convenience of having it on the same hardware on my lap.
I pay very little attention to 8 billion or whatever (or even much smaller) models these days and I don't feel like I'm missing much.
If I were going to use this model, I'd be looking to use it more as is the primary chat interface of a larger system, and having it orchestrate & delegate tasks to other places via tool calls. It's not quite as exciting on the surface as a local "do it all" model, but it does enable some pretty neat use-cases, IMO.
I'm imagining a local agent that is super low latency, works entirely offline, and capable of queuing up complex tasks for larger/smarter cloud agents which execute them asynchronously.
Humans train "thinking methodology" first and then know how to use it while accessing data and to build knowledge.
Humans do not memorize at once all text in existence, that's totally stupid.
Already thinking humans specialize in disciplines: math, chemistry, IT, cooking, etc while still using new data.
All of that computing is local- on the LAN of the brain.
So if some "agents" wants to help then there is zero need for computation outside of home/corporation/car local area network.
Licenses ??