LOCAL_AI_STACK
v2.3 · 2026-08-09
← back to models

LFM2.5 230M

lfm2-5-230m
TEXT-CHAT

LFM2.5 230M is a 0.23B text chat model profile for local AI planning. This page records provider, license, context, quantization, hardware-fit estimates, setup hints, source links, and caveats so readers do not choose a model by name alone.

PROVIDER
Liquid AI
FAMILY
LFM2.5
MODEL TYPE
text-chat
PARAMETERS
0.23B
MODALITIES
text
ARCHITECTURE
Hybrid decoder with gated short-convolution and grouped-query-attention blocks
CONTEXT WINDOW
32,768 tokens
TRAINING TOKENS
19.0T
RELEASE DATE
2026-06-25
LICENSE & LINKS
LFM Open License v1.0license text
GGUF REPOSITORIES
Official GGUF repository for llama.cpp-compatible local use.
SETUP HINTS
LM STUDIO
Import a compatible official artifact and verify the selected runtime.
LLAMA.CPP
Use the official GGUF repository with a current llama.cpp build.
RAM / VRAM ESTIMATES
MIN RAM
4 GB
COMFORTABLE RAM
8 GB
These are conservative local-inference estimates, not official hardware requirements.
HARDWARE FIT
CPU only (no GPU)
yes
8 GB RAM
comfortable
16 GB RAM
comfortable
32 GB RAM
comfortable
8 GB VRAM
not required
12 GB VRAM
not required
24 GB VRAM
not required
Apple Silicon (unified memory)
comfortable
Hardware fit values are conservative local-inference estimates based on GGUF size plus runtime and KV-cache overhead. Actual requirements depend on context length, quantization, runtime, GPU offload, and other running apps.
BEST FOR
·Structured extraction, tool routing, and lightweight on-device agent components.
AVOID IF
·You need advanced reasoning, code generation, or creative writing.
CAVEATS
·Vendor throughput figures have not been reproduced by LocalLLMGuide.com.
·The custom license limits commercial use for entities at or above $10 million in annual revenue.
·The official card does not recommend this model for reasoning-heavy math, code generation, or creative writing.
SOURCE URLS
FIELD EVIDENCE
LANGUAGES
EnglishArabicChineseFrenchGermanItalianJapaneseKoreanPortugueseSpanish
CAPABILITIES
tool calling
← DeepSeek-V3.2LFM2.5 Embedding 350M →