lfm2
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.
6 downloadable Ollama tags for lfm2. Download size is listed when published. Weights are $0; hardware and operating costs are additional.
| Provider | Plan | Operating cost | Download size | Source date |
|---|---|---|---|---|
| lfm2 | lfm2:24b | Total unknown Weights download $0.00 Hardware and operating costs are not priced. | 14.0 GB download | 2026-09-18 |
| lfm2 | lfm2:24b-a2b | Total unknown Weights download $0.00 Hardware and operating costs are not priced. | 14.0 GB download | 2026-09-18 |
| lfm2 | lfm2:24b-bf16 | Total unknown Weights download $0.00 Hardware and operating costs are not priced. | 48.0 GB download | 2026-09-18 |
| lfm2 | lfm2:24b-q4_K_M | Total unknown Weights download $0.00 Hardware and operating costs are not priced. | 14.0 GB download | 2026-09-18 |
| lfm2 | lfm2:24b-q8_0 | Total unknown Weights download $0.00 Hardware and operating costs are not priced. | 25.0 GB download | 2026-09-18 |
| lfm2 | lfm2:latest | Total unknown Weights download $0.00 Hardware and operating costs are not priced. | 14.0 GB download | 2026-09-18 |