LFM3-1.7B (local)
A tiny local model for agents that call tools, fill in structured output or run on a phone or old laptop. About 1 GB at 4-bit. Not for agentic coding or long multi-step reasoning.
Run it locally
ollama run lfm3 (Ollama 0.40.13 or newer on Linux and Windows), or llama-cli -hf LiquidAI/LFM3-1.7B-GGUF:Q4_K_M- Ollama tag
- lfm3 (1.7b; also 1.7b-mxfp8 and 1.7b-mlx-bf16). Mac only in v0.40.12; Linux and Windows since v0.40.13 (Oct 28, 2026)
- LM Studio name
- Not confirmed in the LM Studio catalog yet (Liquid names LM Studio as a launch partner)
- Download size by quantization
- Ollama 1.7 GB (nvfp4); GGUF Q4_K_M 1.1 GB; Q8_0 1.85 GB; BF16 3.47 GB
- License
- LFM Open License v1.0 (commercial use only under $10M annual revenue)
Memory needed is roughly the download size plus 2 to 4 GB for context. See how to run AI locally.
Specs and prices checked 2026-10-29 against the official page.
Used in these setups
Related skills and personas
Agent skills and bot personas that pair well with LFM3-1.7B (local).
GitHub MCP Server
Read and search issues, branches, files, and pull requests via GitHub API authentication.
GitLab MCP Server
GitLab's official MCP server: merge requests, work items, pipelines, code search and Duo Agent Platform flows, under your own GitLab permissions.
Brave Search MCP Server
Fast privacy-focused web search and local index queries without tracking or ad pollution.
Git Version Control MCP Server
Inspect git log, stage files, read diffs, and create branch commits through protocol tools.
The High-Throughput DBA
Relational integrity, optimal index plans, zero sequential scans, and connection pools.
The Cloud & Edge SRE
Five nines uptime, automated rollback pipelines, low-latency edge deployment, and telemetry.
Want tested skills and personas installed in one command? See the AI Loadout packs, from $9.