GraySoft
Projects Models Compare Cloud benchmarks FAQ Download guIDE →
Model Intelligence Sheet

KoarAI/LFM2.5-350M-Thinking-0004-GGUF overview

🧠 LFM2.5 350M Thinking 0004 GGUF Quantized GGUF binaries of KoarAI/LFM2.5 350M Thinking 0004 for ultra fast local inference on CPU, Vulkan, and Metal via llam…

ggufliquid-ailfm2.5reasoningllama-cppollamaruenlicense:apache-2.0endpoints_compatibleregion:usconversational

Runs locally from ~676.3 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
59
Likes
0
Pipeline
—
Author

Repository Files & Downloads

1 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
LFM2.5-350M-Thinking-0004-f16.ggufGGUFF16676.3 MBDownload

Model Details

Model IDKoarAI/LFM2.5-350M-Thinking-0004-GGUF
AuthorKoarAI
Pipeline—
Licenseapache-2.0
Base model—
Last modified2026-08-31T14:01:09.000Z

Model README

---

license: apache-2.0

language:

  • ru
  • en

tags:

  • liquid-ai
  • lfm2.5
  • reasoning
  • gguf
  • llama-cpp
  • ollama

---

🧠 LFM2.5-350M-Thinking-0004-GGUF

Quantized GGUF binaries of KoarAI/LFM2.5-350M-Thinking-0004 for ultra-fast local inference on CPU, Vulkan, and Metal via llama.cpp.

📦 Files Available

  • LFM2.5-350M-Thinking-0004-f16.gguf (~709 MB) — Full Precision FP16 baseline.

🚀 Quick Start (llama.cpp)

llama-cli -m LFM2.5-350M-Thinking-0004-f16.gguf -p "<|im_start|>user\nСколько будет 17 * 19?\n<|im_end|>\n<|im_start|>assistant\n" -n 512 --temp 0.6

Run KoarAI/LFM2.5-350M-Thinking-0004-GGUF with guIDE

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE → · Browse 524k+ models · Compare models

Source: Hugging Face · Compare models