LiquidAI
A United States AI startup focused on developing generative AI models, with an emphasis on efficient liquid neural network architectures.
Timeline (last 90 days)
LiquidAI released the new LFM2.5-24B-A2B model.
LiquidAI's d1-omni-600M runs on CPU, and all 18 newly added blocks were false positives, with recall remaining at the upstream 0.319.
LiquidAI's LFM2.5-24B-A2B uses sparse MoE to activate only 2B per token out of 24B total parameters, extends context from 32K to 131K, runs at 184 tok/s on an Apple M5 Max, and reaches aggregate throughput of up to 21K tok/s on a single NVIDIA B300.
LiquidAI/d1-3B has accumulated 10,843 downloads on Hugging Face since its release.
All Facts (4)
LiquidAI's d1-omni-600M runs on CPU, and all 18 newly added blocks were false positives, with recall remaining at the upstream 0.319.
50%UnverifiedLiquidAI's LFM2.5-24B-A2B uses sparse MoE to activate only 2B per token out of 24B total parameters, extends context from 32K to 131K, runs at 184 tok/s on an Apple M5 Max, and reaches aggregate throughput of up to 21K tok/s on a single NVIDIA B300.
50%UnverifiedLiquidAI/d1-3B has accumulated 10,843 downloads on Hugging Face since its release.
50%UnverifiedLiquidAI released the new LFM2.5-24B-A2B model.
50%