Skip to content
localmodel.run

Text model · dots.llm

DO dots.llm1: RAM and VRAM requirements

dots.llm1 needs about 98.1 GB to run at Q4_K_M (the Q4_K_M GGUF file is ~94.4 GB to download; KV cache and overhead add the rest), or about 155.6 GB at Q8_0. The lightest hardware that runs it is Apple M3 Ultra (256GB).

dots.llm family · 142B params (Mixture-of-Experts: activates only 14B of 142B params per token, so generation is faster than the total size suggests) · released Jun 2025.

Shopping for hardware? See what runs dots.llm1 →

License MIT · Commercial OK ↓ 2.3K/mo ♥ 179 on HuggingFace
Q4_K_M GGUF
94.4 GB
Q8_0 GGUF
151.9 GB
Memory @ Q4 (4k)
~98.1 GB
Context
32 k

Q4_K_M and Q8_0 are GGUF download (file) sizes. Memory @ Q4 includes the KV cache and overhead, so it runs larger than the file on disk.

Will it run on your device?

dots.llm1 runs on 1 of 40 tracked devices at Q4_K_M.

1 run well 0 tight fit 39 too small
Fit check Q4_K_M
No, not enough memory
needs 98.1 GB usable 10.5 GB
See the full breakdown

Memory breakdown

Weights (Q4_K_M)94.4 GB
+
KV cache (4k)2.9 GB
+
Overhead0.8 GB
=
Total98.1 GB

How context length changes it

4k context ~98.1 GB 32k context ~118.1 GB 128k context ~186.7 GB

Longer context grows the KV cache, which for dots.llm1 is sized by its 14B active params, not the full 142B. It needs ~98.1 GB at 4k and ~186.7 GB at 128k.

Quantization sizes

GGUF quantson disk
Quantization Size on disk
Q2_K 59.5 GB est
Q3_K_M 69.4 GB est
Q4_K_M (default) 94.4 GB
Q5_K_M 101.2 GB est
Q6_K 116.4 GB est
Q8_0 151.9 GB
FP16 286 GB

Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.

Run it

llama.cpp
$ llama-cli -hf mradermacher/dots.llm1.inst-GGUF:Q4_K_M
LM Studio
$ lms get mradermacher/dots.llm1.inst-GGUF

Which devices can run dots.llm1?

Apple Silicon Macs

RAM-only laptops

iPhone & iPad

Android

NVIDIA GPUs

AMD GPUs

Same job, different size

Pick by what fits your memory: step down to free up VRAM, or step up if you have headroom.

Next size down Nemotron 3 Super 120B-A12B 120B Next size up Qwen3 235B A22B 235B

Similar models

FAQ

How much VRAM or RAM does dots.llm1 need?

At Q4_K_M, dots.llm1 needs about 98.1 GB (weights ~94.4 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~155.6 GB.

What is the Q4_K_M GGUF file size of dots.llm1?

The Q4_K_M GGUF file is about 94.4 GB to download, and the Q8_0 GGUF is about 151.9 GB. That is the weights file on disk; to run it you also need room for the KV cache and overhead, so budget ~98.1 GB of memory at Q4_K_M.

Can dots.llm1 run on a laptop?

dots.llm1 is large; you need a high-memory Mac or multi-GPU setup at Q4_K_M.

Is dots.llm1 cheaper to run because it is a MoE model?

It is faster, not lighter. dots.llm1 activates only 14B of 142B params per token (so it runs quickly), but all experts must stay in memory, so it still needs memory for the full 142B.

Can I use dots.llm1 commercially?

Yes. dots.llm1 is licensed MIT, which permits commercial use.

Understand the numbers

Short guides to the ideas behind dots.llm1's memory and quant figures.

Rednote's open MoE: 142B total / 14B active, MIT.

Sources

Last validated 2026-08-03. Memory figures are estimates. See methodology.