IC

ICONN 1

IconnGGUF by gabriellarsonUpdated 2025-06-19
Download on Hugging Face
Parameters84B
Context32K
Layers40
Architecturellama
Smallest quant17.4 GB
Quick answer

ICONN 1 needs about 52.7 GB of memory at Q4_K_M (a 50.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 24 quants, from 17.4 GB (IQ1_S) to 89.2 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 5.4 GB.

ICONN 1 is a 84B-parameter Iconn model in GGUF format, quantized by gabriellarson. Its trained context length is 32K tokens. The repository has 24 quant files, from 17.4 GB (IQ1_S) to 89.2 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S17.4 GB 19.2 GB
Too big Get file →
IQ1_M19.2 GB 21.1 GB
Too big Get file →
IQ2_XXS22.3 GB 24.1 GB
Too big Get file →
IQ2_XS24.8 GB 26.6 GB
Too big Get file →
IQ2_S25.3 GB 27.1 GB
Too big Get file →
IQ2_M27.7 GB 29.6 GB
Too big Get file →
Q2_K_S28.7 GB 30.6 GB
Too big Get file →
Q2_K30.8 GB 32.6 GB
Too big Get file →
IQ3_XXS32.4 GB 34.2 GB
Too big Get file →
IQ3_XS34.4 GB 36.3 GB
Too big Get file →
Q3_K_S36.4 GB 38.2 GB
Too big Get file →
IQ3_S36.4 GB 38.2 GB
Too big Get file →
IQ3_M36.9 GB 38.8 GB
Too big Get file →
Q3_K_M40.2 GB 42.1 GB
Too big Get file →
Q3_K_L43.5 GB 45.4 GB
Too big Get file →
IQ4_XS44.8 GB 46.7 GB
Too big Get file →
Q4_047.6 GB 49.5 GB
Too big Get file →
Q4_K_S47.8 GB 49.7 GB
Too big Get file →
Q4_K_M50.9 GB 52.7 GB
Too big Get file →
Q5_K_S57.8 GB (2 parts) 59.7 GB
Too big Get file →
Q5_058.0 GB 59.9 GB
Too big Get file →
Q5_K_M59.6 GB (2 parts) 61.5 GB
Too big Get file →
Q6_K68.9 GB (2 parts) 70.7 GB
Too big Get file →
Q8_089.2 GB (2 parts) 91.1 GB
Too big Get file →

What GPU do you need for ICONN 1?

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: the best quant that fits is IQ1_M (19.2 GB file, about 21.1 GB in use).
  • 32 GB VRAM: the best quant that fits is Q2_K_S (28.7 GB file, about 30.6 GB in use).
  • 48 GB VRAM: the best quant that fits is IQ4_XS (44.8 GB file, about 46.7 GB in use).

Which ICONN 1 quant should you choose?

Q4_K_M is the usual starting point for ICONN 1: it balances file size and quality well. Here it is a 50.9 GB file.

The Q8_0 file is 89.2 GB, about 1.8 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 17.4 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

Architecturellama
Parameters84B (approximate, mixture of experts)
Layers40
Trained context32,768 tokens
KV cache (fp16)160 KB per token

Source

Quantized bygabriellarson
FamilyIconn
Licenseother
Downloads (HF)245
Last updated2025-06-19

Frequently asked questions about ICONN 1

How much VRAM does ICONN 1 need?

About 52.7 GB at Q4_K_M with an 8K context window. That is the 50.9 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 ICONN 1 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 19.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of ICONN 1 GGUF?

It depends on the quant: IQ1_S is 17.4 GB, IQ1_M is 19.2 GB, IQ2_XXS is 22.3 GB, IQ2_XS is 24.8 GB, IQ2_S is 25.3 GB, IQ2_M is 27.7 GB and 18 more.

What context length does ICONN 1 support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.3 GB at 8K and 5.4 GB at 32K.

Where can I download ICONN 1 GGUF?

The files are on Hugging Face in the gabriellarson/ICONN-1-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 ICONN 1 use?

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

Similar models

Browse all →

More from gabriellarson

All by gabriellarson →

More Iconn models

All Iconn →

Other large models

All large models →