LFM2.5-350M
A lightweight 350M-parameter language model released by Liquid AI, supporting ONNX browser builds and targeting on-device inference scenarios.
Timeline (last 90 days)
Candidate small models include SmolLM2 (135M and 360M), Gemma 3 270M, LFM2.5-350M, Qwen2.5-0.5B, and Qwen3-0.6B, all of which have ONNX browser builds.
In an evaluation of 30 answerable questions, off-the-shelf small models under 350MB achieved faithfulness (all assertions can be found in the passages) at most 3 times.
All Facts (2)
Candidate small models include SmolLM2 (135M and 360M), Gemma 3 270M, LFM2.5-350M, Qwen2.5-0.5B, and Qwen3-0.6B, all of which have ONNX browser builds.
50%UnverifiedIn an evaluation of 30 answerable questions, off-the-shelf small models under 350MB achieved faithfulness (all assertions can be found in the passages) at most 3 times.
50%