LFM3 1.7B VRAM requirements
Liquid AI · 1.73B params · LFM Open License v1.0. Estimates, not vendor numbers.
LFM3 1.7B (1.73B parameters) needs about 2.5 GB of VRAM or unified memory at Q4, 2.7 GB at Q5 and 3.3 GB at Q8 (weights plus about 1.5 GB headroom, our estimate). The smallest common setup that runs it is 8GB GPU (RTX 4060).
| Quant | Download | Memory needed |
|---|---|---|
| Q4 | 1.0 GB | 2.5 GB |
| Q5 | 1.2 GB | 2.7 GB |
| Q8 | 1.8 GB | 3.3 GB |
Will it run on my machine?
| Setup | Runs? | Speed est. |
|---|---|---|
| 8GB GPU (RTX 4060) | Yes, Q8 | ~89 tok/s |
| 12GB GPU (RTX 3060 / 4070) | Yes, Q8 | ~165 tok/s |
| 16GB GPU (RTX 4060 Ti 16GB) | Yes, Q8 | ~94 tok/s |
| 24GB GPU (RTX 4090) | Yes, Q8 | ~329 tok/s |
| 32GB GPU (RTX 5090) | Yes, Q8 | ~585 tok/s |
| Mac 16GB (M4) | Yes, Q8 | ~39 tok/s |
| Mac 48GB (M4 Pro) | Yes, Q8 | ~89 tok/s |
| Mac 128GB (M4 Max) | Yes, Q8 | ~178 tok/s |
| PC, 32GB RAM, no GPU | Yes, Q8 | ~20 tok/s |
Best quant that fits each setup, with estimated generation speed from memory bandwidth. Real speed depends on the runtime, context length and settings.
Small text model for local agents, tool calling and structured output, 256K context. Posted Oct 23, 2026. Liquid AI says it takes about 1 GB of memory at 4-bit and doesn't recommend it for agentic coding or complex multi-step reasoning.
FAQ
How much VRAM does LFM3 1.7B need?
About 2.5 GB at Q4, 2.7 GB at Q5 and 3.3 GB at Q8, including about 1.5 GB headroom. These are estimates from the parameter count on the official model card.
Can LFM3 1.7B run on 8GB of VRAM?
Yes, at Q8.
Can LFM3 1.7B run on a Mac?
Yes. Mac 16GB (M4), Mac 48GB (M4 Pro), Mac 128GB (M4 Max) can run it.
Which quant of LFM3 1.7B should I use?
Use the highest quant that fits with room to spare: Q8 is near full quality, Q5 is a good balance, and Q4 is the smallest that usually keeps quality acceptable.
Similar-size models
Ministral 3 3B (3B) · Phi-4 mini 3.8B (3.8B) · Qwen3 4B (4B) · Gemma 4 E4B (8B) · Ministral 3 8B (8B) · Llama 3.1 8B Instruct (8B)