Try LFM •
Docs •
Discord
LFM2.5-24B-A2B-FP8
FP8 version of LFM2.5-24B-A2B for GPU inference.
Linear weights are quantized to FP8 with 128×128 blocks and dynamic activation quantization. The embedding, MoE routing gates and convolution weights are kept in their original precision.
Find more details in the original model card: https://huggingface.co/LiquidAI/LFM2.5-24B-A2B
Use with SGLang
python -m sglang.launch_server --model-path LiquidAI/LFM2.5-24B-A2B-FP8
License
This model is released under the LFM 1.0 License.