Always put "Lower is better" or "higher is better" in benchmarks. Not everyone knows what your numbers mean.
lend000
I can't imagine who is using something like this for agentic coding, but I see exciting opportunities on the horizon when we can have hundreds of reasonably rational and conversational agents working on local machines to simulate emergent behavior (simulating crowds, markets, ecosystems, game NPCs, etc.)
show comments
Gecko4072
These LiquidAI models have never worked well for me in practice.
show comments
lostmsu
It's not even competitive with 2x sized Qwen 4B.
Why is Qwen3.5 2B not in the table?
0xbadcafebee
LFM's training/post-training is famously different than other models. They target reliable operation of tiny models in ways other model families don't (they aren't just scaling a larger model to a smaller size). If you're looking for good performance out of tiny models, LFM has the most advanced design.
Note how they're much smaller than all other models in the comparison yet match or exceed them. This is for 2.6B params, but they have models as small as 230M. Nobody else designs models that small.
show comments
harshshah212003
Will this work in i3/i5 laptops?
show comments
GaggiX
The model is cool but I would prefer if people do not editorialize the titles on their HN submissions.
Always put "Lower is better" or "higher is better" in benchmarks. Not everyone knows what your numbers mean.
I can't imagine who is using something like this for agentic coding, but I see exciting opportunities on the horizon when we can have hundreds of reasonably rational and conversational agents working on local machines to simulate emergent behavior (simulating crowds, markets, ecosystems, game NPCs, etc.)
These LiquidAI models have never worked well for me in practice.
It's not even competitive with 2x sized Qwen 4B.
Why is Qwen3.5 2B not in the table?
LFM's training/post-training is famously different than other models. They target reliable operation of tiny models in ways other model families don't (they aren't just scaling a larger model to a smaller size). If you're looking for good performance out of tiny models, LFM has the most advanced design.
Note how they're much smaller than all other models in the comparison yet match or exceed them. This is for 2.6B params, but they have models as small as 230M. Nobody else designs models that small.
Will this work in i3/i5 laptops?
The model is cool but I would prefer if people do not editorialize the titles on their HN submissions.