{"v":1,"id":"model:hf:LiquidAI/LFM2.5-230M-GGUF","slug":"model-liquidai-lfm2-5-230m-gguf","kind":"model","category":"llm","title":"LFM2.5-230M-GGUF","summary":"LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment. It sets a new standard in terms of quality, speed, and memory efficiency.","source":{"provider":"hf","ref":"LiquidAI/LFM2.5-230M-GGUF","url":"https://huggingface.co/LiquidAI/LFM2.5-230M-GGUF","rev":"03502067c64ce32ac4fe87b0cec0310a1a13d3e9","fetchedAt":"2026-10-02T21:00:08.537Z","etag":"W/\"270a-brH/u2u71/WjKuKTE5qMBM9wzRo\""},"author":{"name":"LiquidAI","url":"https://huggingface.co/LiquidAI"},"license":{"spdx":null,"raw":"other","open":null,"note":"custom license: read it at the source before installing"},"metrics":{"downloads":534666,"downloadsWeek":327838,"likes":110,"takenAt":"2026-10-02T21:00:08.537Z"},"tags":["gguf","safetensors","liquid","lfm2.5","llama.cpp","text-generation","en","ar","zh","fr","de","ja","ko","es","pt","it","endpoints_compatible","conversational"],"pipeline":"text-generation","links":{"github":"ggml-org/llama.cpp","npm":"node-llama-cpp"},"updatedAt":"2026-09-22T20:41:15.000Z","collectedAt":"2026-10-02T21:00:08.537Z","review":{"numbers":["534,666 downloads on Hugging Face","110 likes","license other","0.1 GB for LFM2.5-230M-Q4_K_M.gguf","327,838 npm downloads a week for node-llama-cpp","latest node-llama-cpp@3.22.1"],"log":null},"trust":"unlabeled","health":{"status":"alive","checkedAt":"2026-10-02T21:00:08.537Z","http":200},"description":"Try LFM •\n    Docs •\n    LEAP •\n    Discord\n\n# LFM2.5-230M-GGUF\n\nLFM2 is a new generation of hybrid models developed by [Liquid AI](https://www.liquid.ai/), specifically designed for edge AI and on-device deployment. It sets a new standard in terms of quality, speed, and memory efficiency.\n\nFind more details in the original model card: https://huggingface.co/LiquidAI/LFM2.5-230M\n\n## 🏃 How to run LFM2.5\n\nExample usage with [llama.cpp](https://github.com/ggml-org/llama.cpp):\n\n```\nllama-cli -hf LiquidAI/LFM2.5-230M-GGUF --conversation \\\n    --temp 0.1 --top-k 50 --repeat-penalty 1.05\n```\n\n## QAD Q4_0 GGUF\n\nThe Quantization-Aware Distillation (QAD) checkpoint is available as\n[`LFM2.5-230M-QAD-Q4_0.gguf`](https://huggingface.co/LiquidAI/LFM2.5-230M-GGUF/blob/main/LFM2.5-230M-QAD-Q4_0.gguf).\n\nThis is distinct from the post-training-quantized `LFM2.5-230M-Q4_0.gguf`;\nboth use the GGUF Q4_0 format.\n\nExample usage with [llama.cpp](https://github.com/ggml-org/llama.cpp):\n\n```shell\nllama-cli -hf LiquidAI/LFM2.5-230M-GGUF \\\n  --hf-file LFM2.5-230M-QAD-Q4_0.gguf \\\n  -p \"What is C. elegans?\"\n```\n\n## QAD source weights (safetensors)\n\nThe original FP32 QAD source checkpoint is available in\n[`qad/`](https://huggingface.co/LiquidAI/LFM2.5-230M-GGUF/tree/main/qad), with its model config,\ntokenizer, generation defaults, and the same chat template as the released QAD GGUF.\nIt can be loaded in Transformers by passing `subfolder=\"qad\"`:\n\n```python\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nrepo_id = \"LiquidAI/LFM2.5-230M-GGUF\"\ntokenizer = AutoTokenizer.from_pretrained(repo_id, subfolder=\"qad\")\nmodel = AutoModelForCausalLM.from_pretrained(\n    repo_id, subfolder=\"qad\", dtype=\"auto\", device_map=\"auto\"\n)\n\ninputs = tokenizer.apply_chat_template(\n    [{\"role\": \"user\", \"content\": \"What is 2 + 2?\"}],\n    tokenize=True, add_generation_prompt=True, return_dict=True, return_tensors=\"pt\",\n).to(model.device)\noutputs = mode…\n\nSource: https://huggingface.co/LiquidAI/LFM2.5-230M-GGUF","install":{"kind":"model","hfId":"LiquidAI/LFM2.5-230M-GGUF","gated":false,"format":"gguf","files":[{"name":"LFM2.5-230M-BF16.gguf","size":461884256,"quant":"BF16","sha256":"9a47cffa8c86d071e4cdb2adf6861251fecd6191c226c88607a193a0d9cc5a38"},{"name":"LFM2.5-230M-F16.gguf","size":461884256,"quant":"F16","sha256":"4d364976c7ae1b85bd380f743155aa2d532f7a10291beaa6b27a7d6c9b10527f"},{"name":"LFM2.5-230M-Q4_0.gguf","size":149080928,"quant":"Q4_0","sha256":"430fbec5b1b355e9bb12cd0638c9f2a8f21fedd6eafb4103e42c7e88887daa73"},{"name":"LFM2.5-230M-Q4_K_M.gguf","size":153406304,"quant":"Q4_K_M","sha256":"7bbd90384d3deffe4c646ec9643b212802d32d4ce417c90a1ec9282100650062"},{"name":"LFM2.5-230M-Q5_K_M.gguf","size":171625312,"quant":"Q5_K_M","sha256":"65bceea824d701506de53f02d7276578a64706ad9c17f00d6e8917bf491de6db"},{"name":"LFM2.5-230M-Q6_K.gguf","size":190983008,"quant":"Q6_K","sha256":"17a451304bfdfa7098902d5e8e0122440bbfc585f722bd1475435cbc5bab0219"},{"name":"LFM2.5-230M-Q8_0.gguf","size":246598496,"quant":"Q8_0","sha256":"855be85429300602eda72958547614703541b7d6dd965a8f8f6052b85a7aa935"},{"name":"LFM2.5-230M-QAD-Q4_0.gguf","size":149081056,"quant":"Q4_0","sha256":"e75f83268de11b2a1bcfab5f3b5c5c0c97569ddbbc0990aad88437e45b8ba292"},{"name":"qad/model.safetensors","size":918787352,"sha256":"1a55fc507f92944f735e36181c6fb2e61418c49e695b98938a4ac74d45d73f0e"}],"totalBytes":2903330968,"suggestedFile":"LFM2.5-230M-Q4_K_M.gguf","requirements":{"ramGb":1,"diskBytes":153406304,"note":"estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models"},"runWith":["llama.cpp","ollama"],"command":"lsh models install hf:LiquidAI/LFM2.5-230M-GGUF"}}