DO

Dots.Llm1.Inst

DotsGGUF by bartowskiUpdated 2025-06-16
Download on Hugging Face
Parameters143B
Context32K
Layers62
Architecturedots1
Smallest quant47.3 GB
Quick answer

Dots.Llm1.Inst needs about 104.9 GB of memory at Q4_K_M (a 96.1 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 26 quants, from 47.3 GB (IQ1_S) to 151.8 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 33.3 GB.

Dots.Llm1.Inst – a 143B parameter MoE model with 14B active parameters

Dots.Llm1.Inst is a 143B parameter MoE model with 14B activated parameters. It supports a 32,768 token context length and is licensed under MIT. The model is available in multiple GGUF quantizations.

Good at

  • Large context handling
  • High parameter efficiency
  • Instruction tuning

Limits

  • No vision support
  • Requires large VRAM for high‑quant models
  • No language list provided

Best for

  • Long‑form text generation
  • Complex reasoning tasks
  • Research on MoE architectures

Who it’s for. Researchers and developers needing a large‑scale MoE model with a long context window. The model is suitable for experimentation and fine‑tuning in llama.cpp or LM Studio environments.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S47.3 GB 56.1 GB
Too big Get file →
IQ1_M48.1 GB 57.0 GB
Too big Get file →
IQ2_XXS50.5 GB (2 parts) 59.3 GB
Too big Get file →
IQ2_XS53.9 GB (2 parts) 62.7 GB
Too big Get file →
IQ2_S54.1 GB (2 parts) 62.9 GB
Too big Get file →
IQ2_M57.6 GB (2 parts) 66.5 GB
Too big Get file →
Q2_K60.2 GB (2 parts) 69.1 GB
Too big Get file →
Q2_K_L60.8 GB (2 parts) 69.7 GB
Too big Get file →
IQ3_XXS65.1 GB (2 parts) 73.9 GB
Too big Get file →
IQ3_XS66.3 GB (2 parts) 75.1 GB
Too big Get file →
Q3_K_S69.6 GB (2 parts) 78.4 GB
Too big Get file →
IQ3_M72.5 GB (2 parts) 81.3 GB
Too big Get file →
Q3_K_M72.5 GB (2 parts) 81.3 GB
Too big Get file →
Q3_K_L73.7 GB (2 parts) 82.5 GB
Too big Get file →
Q3_K_XL74.3 GB (2 parts) 83.1 GB
Too big Get file →
IQ4_XS79.0 GB (2 parts) 87.8 GB
Too big Get file →
IQ4_NL81.9 GB (3 parts) 90.7 GB
Too big Get file →
Q4_083.1 GB (3 parts) 92.0 GB
Too big Get file →
Q4_K_S88.9 GB (3 parts) 97.7 GB
Too big Get file →
Q4_190.4 GB (3 parts) 99.3 GB
Too big Get file →
Q4_K_M96.1 GB (3 parts) 104.9 GB
Too big Get file →
Q4_K_L96.6 GB (3 parts) 105.4 GB
Too big Get file →
Q5_K_S102.1 GB (3 parts) 110.9 GB
Too big Get file →
Q5_K_M109.4 GB (3 parts) 118.2 GB
Too big Get file →
Q6_K128.9 GB (4 parts) 137.7 GB
Too big Get file →
Q8_0151.8 GB (4 parts) 160.6 GB
Too big Get file →

For 8 GB VRAM use IQ1_S (47.29 GB file, 56.1 GB memory). For 16 GB VRAM use Q4_K_M (96.12 GB file, 104.9 GB memory). For 32 GB VRAM use Q8_0 (151.8 GB file, 160.6 GB memory). Memory estimates are approximate.

What GPU do you need for Dots.Llm1.Inst?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Which Dots.Llm1.Inst quant should you choose?

Q4_K_M is the usual starting point for Dots.Llm1.Inst: it balances file size and quality well. Here it is a 96.1 GB file.

The Q8_0 file is 151.8 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is IQ1_S at 47.3 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturedots1
Parameters143B (approximate, mixture of experts)
Layers62
Trained context32,768 tokens
KV cache (fp16)992 KB per token

Source

Quantized bybartowski
FamilyDots
Licensemit
Downloads (HF)573
Last updated2025-06-16

Frequently asked questions about Dots.Llm1.Inst

How much VRAM does Dots.Llm1.Inst need?

About 104.9 GB at Q4_K_M with an 8K context window. That is the 96.1 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Dots.Llm1.Inst on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 56.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Dots.Llm1.Inst GGUF?

It depends on the quant: IQ1_S is 47.3 GB, IQ1_M is 48.1 GB, IQ2_XXS is 50.5 GB, IQ2_XS is 53.9 GB, IQ2_S is 54.1 GB, IQ2_M is 57.6 GB and 20 more.

What context length does Dots.Llm1.Inst support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 8.3 GB at 8K and 33.3 GB at 32K.

Where can I download Dots.Llm1.Inst GGUF?

The files are on Hugging Face in the bartowski/rednote-hilab_dots.llm1.inst-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Dots.Llm1.Inst use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run IQ1_S on 8 GB VRAM?

Yes, IQ1_S requires about 56.1 GB memory at 8k context, so it will not fit in 8 GB VRAM.

Does this model support Chinese?

Yes, the model supports English and Chinese.

Is the model licensed for commercial use?

The model is licensed under MIT, which allows commercial use.

Other GGUFs of dots.llm1.inst

Similar models

Browse all →

More from bartowski

All by bartowski →

More Dots models

All Dots →

Other large models

All large models →