GraySoft
Model Comparison

liquidai/lfm2-24b-a2b-ggufvsunsloth/nvidia-nemotron-3-super-120b-a12b-gguf

Side-by-side comparison of liquidai/lfm2-24b-a2b-gguf and unsloth/nvidia-nemotron-3-super-120b-a12b-gguf: downloads, license, context length, tasks, and benchmarks.

liquidai/lfm2-24b-a2b-gguf

LiquidAI · text-generation

LFM2 is a family of hybrid models designed for on-device deployment. LFM2-24B-A2B is the largest model in the family, scaling the architecture to 24 billion parameters while keeping inference efficient. !image Find more information about LFM2-24B-A2B in our blog post.

unsloth/nvidia-nemotron-3-super-120b-a12b-gguf

unsloth · text-generation

![](./accuracy_chart.png)

Side-by-side Specifications

liquidai/lfm2-24b-a2b-ggufunsloth/nvidia-nemotron-3-super-120b-a12b-gguf
AuthorLiquidAIunsloth
Pipeline Tasktext-generationtext-generation
Librarytransformers
Downloads56,74879,811
Likes122113
LicenseUnknownUnknown
Context Length
Created2026-02-172026-03-06
Last Modified2026-03-302026-03-20
Tags
transformersggufliquidlfm2edgemoellama.cpptext-generationenar
ggufnvidiapytorchnemotron-3latent-moemtpunslothtext-generationenfr

View full details: liquidai/lfm2-24b-a2b-gguf · unsloth/nvidia-nemotron-3-super-120b-a12b-gguf