text model · LFM · macOS
Can I run LFM2 350M on Apple M4 Pro (48GB)?
Yes. LFM2 350M runs on Apple M4 Pro (48GB) at Q4_K_M (~1.4 GB of ~32 GB usable).
Runs at Q4_K_M using ~1.4 GB of ~32 GB usable. You have room for FP16 for higher quality.
That figure is at a 4k context and moves about ±15% as context length changes. Apple M4 Pro (48GB) leaves ~30.6 GB of headroom, room to step up to FP16 for higher quality.
- Q4_K_M needed
- ~1.4 GB
- Usable on device
- ~32 GB
- Device memory
- 48 GB
- Best quant
- Q4_K_M
Which quant fits
Running cost · estimate
- Power draw
- ~140 W
- Electricity / 1M tokens
- ~$0.01
- Pays for itself after
- ~4,896M tok
At ~$0.15/kWh and the estimated ~485 tok/s, a million generated tokens costs about $0.01 in electricity. Versus a hosted API at ~$0.5 per million tokens, the ~$2,399 Apple M4 Pro (48GB) pays for itself after roughly 4,896 million tokens, so local hardware is mostly a fixed cost, not a per-token one. TDP is the peak draw, so this is an upper bound. Assumptions.
Run it
Pick your tool. All 2 load the same Q4_K_M weights.
llama-cli -hf LiquidAI/LFM2-350M-GGUF:Q4_K_M lms get LiquidAI/LFM2-350M-GGUF How to run it
On macOS use LM Studio (Polished GUI, ships MLX on Apple Silicon, one-click model downloads.).
- Parameters
- 0.354B
- Q4_K_M size
- 0.45 GB
- Q8_0 size
- 0.35 GB
- Context
- 128k
- Memory
- 48 GB unified
- Usable for weights
- ~32 GB
- Power draw
- ~140 W
- Best runtime
- Ollama (MLX backend) / MLX direct
You could also run
Run LFM2 350M on other hardware
FAQ
Can Apple M4 Pro (48GB) run LFM2 350M?
Yes. LFM2 350M runs on Apple M4 Pro (48GB) at Q4_K_M (~1.4 GB of ~32 GB usable).
How much memory does LFM2 350M need?
Apple M4 Pro (48GB) has room to spare. At Q4_K_M the weights are ~0.45 GB; with KV cache and runtime overhead, budget ~1.4 GB at a 4k context.
What is the best tool to run LFM2 350M on macOS?
LM Studio for a simple setup; mlx-lm for the most speed. vLLM is NOT a Mac tool, it is a CUDA/Linux serving engine. Unified memory is not a fixed VRAM slice; ~70% is usable for weights.
Embed this
A live badge for your model card or README, updated as the data is. [](https://localmodel.run/can-i-run/lfm2-350m/apple-m4-pro-48gb) Sources
- apple.com/newsroom/2024/10/apple-introduces-m4-pro-and-m4-max
- apple.com/newsroom/2024/10/new-macbook-pro-features-m4-family-of-chips-and-apple-intelligence
- arxiv.org
- blog.peddals.com
- github.com/ml-explore
- github.com/raullenchai
- huggingface.co/LiquidAI/LFM2-350M
- huggingface.co/LiquidAI/LFM2-350M-GGUF
- liquid.ai
- lmstudio.ai
- support.apple.com/en-us/103253
- support.apple.com/en-us/121553
- support.apple.com/en-us/121555
Weights are measured from GGUF files; KV cache and overhead are computed, so totals can vary ~15% with context and runtime. Any tok/s is a bandwidth estimate. See methodology.