license, license_name, license_link, base_model, language, pipeline_tag, tags, library_name
| license | license_name | license_link | base_model | language | pipeline_tag | tags | library_name | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| other | lfm1.0 | LICENSE |
|
|
text-generation |
|
transformers |
LiquidAI-LFM2.5-230M-GGUF
LFM2.5-230M is Liquid AI's most compact hybrid model to date, a 230-million-parameter, general-purpose instruction-tuned text model built on the LFM2 architecture with extended pre-training (19T tokens) and reinforcement learning, designed specifically for on-device deployment in the tightest memory and compute budgets. Its 14-layer architecture combines 8 double-gated LIV convolution blocks with 6 GQA blocks, supports a 32,768-token context window across 10 languages, and was distilled from the larger LFM2.5-350M before being refined with multi-stage reinforcement learning, making it well-suited for agentic tasks like tool use and data extraction rather than reasoning-heavy workloads such as advanced math, code generation, or creative writing. It delivers strong edge inference throughput — 213 tok/s decode speed on a Galaxy S25 Ultra and 42 tok/s on a Raspberry Pi 5 — and despite its tiny size, outperforms similarly-scaled competitors like Granite 4.0-350M and LFM2-350M on benchmarks including IFEval (71.71), BFCLv3 (43.26), and Multi-IF (37.70), while supporting native function calling via Pythonic tool calls and ChatML-style chat templates, with deployment options spanning Transformers, vLLM, llama.cpp (GGUF), ONNX, and MLX formats.
Model Files
| File Name | Quant Type | File Size | File Link |
|---|---|---|---|
| LFM2.5-230M.BF16.gguf | BF16 | 462 MB | Download |
| LFM2.5-230M.F16.gguf | F16 | 462 MB | Download |
| LFM2.5-230M.F32.gguf | F32 | 921 MB | Download |
| LFM2.5-230M.Q2_K.gguf | Q2_K | 116 MB | Download |
| LFM2.5-230M.Q3_K_L.gguf | Q3_K_L | 139 MB | Download |
| LFM2.5-230M.Q3_K_M.gguf | Q3_K_M | 134 MB | Download |
| LFM2.5-230M.Q3_K_S.gguf | Q3_K_S | 127 MB | Download |
| LFM2.5-230M.Q4_0.gguf | Q4_0 | 149 MB | Download |
| LFM2.5-230M.Q4_K_M.gguf | Q4_K_M | 153 MB | Download |
| LFM2.5-230M.Q4_K_S.gguf | Q4_K_S | 150 MB | Download |
| LFM2.5-230M.Q5_0.gguf | Q5_0 | 169 MB | Download |
| LFM2.5-230M.Q5_K_M.gguf | Q5_K_M | 172 MB | Download |
| LFM2.5-230M.Q5_K_S.gguf | Q5_K_S | 169 MB | Download |
| LFM2.5-230M.Q6_K.gguf | Q6_K | 191 MB | Download |
| LFM2.5-230M.Q8_0.gguf | Q8_0 | 247 MB | Download |
llama.cpp
LLM inference in C/C++ — https://github.com/ggml-org/llama.cpp