How much memory do I need to run shisa-v2-llama3.1-405b (Q4_K_M)?

The file size is 243 GB (measured). With a 8,192-token context, the estimated memory including the KV cache is 248 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.

Does shisa-v2-llama3.1-405b run in 8 GB, 12 GB, 16 GB or 24 GB of VRAM?

8 GB: no quantization fits; 12 GB: no quantization fits; 16 GB: no quantization fits; 24 GB: no quantization fits (estimated at 8,192 tokens with an f16 KV cache, files from the most-downloaded source).

What is the command to run shisa-v2-llama3.1-405b with Ollama?

Run "ollama run hf.co/DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF:Q4_K_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.

shisa-ai

shisa-v2-llama3.1-405b

Parameters: 405.9B
Max context length: 131,072
Architecture: llama
License: Conditional

Base model: shisa-ai/shisa-v2-llama3.1-405b
License: llama3.1

Memory needed to run Q4_K_M at 8K tokens

248 GB Estimated

File size 243 GB Measured

VRAM it fits in (rough guide)

8 GB does not fit
12 GB does not fit
16 GB does not fit
24 GB does not fit
32 GB does not fit
48 GB does not fit

The smallest, Q2_K, needs an estimated 154 GB at 8K.

Macs cap how much memory the GPU can use, so leave headroom in unified memory.

Data as of Nov 11, 2026

About shisa-v2-llama3.1-405b

shisa-v2-llama3.1-405b is a GGUF model based on shisa-ai/shisa-v2-llama3.1-405b on Hugging Face. GGUF files are published by 2 sources (DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF, shisa-ai/shisa-v2-llama3.1-405b-GGUF). It has about 405.9B parameters, the llama architecture, a maximum context length of 131,072 tokens (values written in the GGUF).

There are 10 quantizations, with file sizes from 107 GB (IQ2_XXS) to 431 GB (Q8_0) (actual sizes on Hugging Face).

Running Q4_K_M (243 GB) with an 8,192-token context needs an estimated 248 GB of memory, and an estimated 261 GB at 32,768 tokens. The smallest, IQ2_XXS (107 GB), needs an estimated 112 GB at 8,192 tokens.

The original model is licensed under Llama 3.1 Community License, which allows commercial use with conditions such as scale or purpose. Check the original license for the terms of use.

File size and estimated memory per quantization

Up to 8 GB
Up to 16 GB
Up to 24 GB
Up to 48 GB
Up to 128 GB
More

File sizes are the actual values on Hugging Face. Memory is estimated for each context length (KV cache f16). The formula is in How memory is calculated.

All memory figures are Estimated (KV cache f16). It should run on a setup with at least this much VRAM (or Mac unified memory). Click a value to see the devices and models that fit that amount (Find by GPU / Mac).

2 sources publish GGUF files of this model. Quantization types and file sizes differ slightly between sources, so there is a table for each source below.

About the "quality guide" (how far a quantization drifts from the original)

For each quantization type, how far the output drifts from the original model, grouped into six levels from values llama.cpp measured on Llama-3-8B.

The value is mean KLD (the difference between the original and quantized model in next-token probabilities); smaller means closer to the original. Quantizations made with an imatrix (importance data) come out slightly smaller. Levels use the value without an imatrix (or with one, for types that only have that).

These values were not measured on this model. Larger models and MoE models tend to drift less with the same type. Distributor-specific mixed types (such as UD-Q4_K_XL) cannot be matched by type name alone, so they show "-".

Quantization	Quality guide	KLD (no imatrix)	KLD (imatrix)	Size (Llama-3-8B)
Q8_0	Nearly identical	0.0014	-	7.96 GiB
Q6_K	Very small	0.0055	-	6.14 GiB
Q5_K_M	Very small	0.0108	-	5.33 GiB
Q5_K_S	Small	0.0166	-	5.21 GiB
Q5_1	Small	0.0180	-	5.65 GiB
Q5_0	Small	0.0222	-	5.21 GiB
Q4_K_M	Small	0.0313	0.0282	4.58 GiB
Q4_K_S	Medium	0.0431	0.0320	4.37 GiB
IQ4_NL	Small	-	0.0357	4.35 GiB
IQ4_XS	Small	-	0.0363	4.14 GiB
Q4_1	Medium	0.0717	-	4.78 GiB
Q4_0	Medium	0.0719	-	4.34 GiB
Q3_K_L	Medium	0.0872	0.0731	4.03 GiB
Q3_K_M	Medium	0.1019	0.0844	3.74 GiB
IQ3_M	Medium	-	0.1025	3.53 GiB
IQ3_S	Medium	-	0.1113	3.42 GiB
IQ3_XS	Large	-	0.1387	3.28 GiB
IQ3_XXS	Large	-	0.1836	3.05 GiB
Q3_K_S	Large	0.2282	0.1998	3.41 GiB
IQ2_M	Large	-	0.3260	2.74 GiB
Q2_K	Very large	0.4451	0.3322	2.96 GiB
Q2_K_S	Very large	-	0.4034	2.96 GiB
IQ2_S	Very large	-	0.4392	2.56 GiB
IQ2_XS	Very large	-	0.5463	2.43 GiB
IQ2_XXS	Very large	-	0.8120	2.24 GiB
IQ1_M	Very large	-	1.3931	2.01 GiB
IQ1_S	Very large	-	2.2546	1.88 GiB

Source: llama.cpp (MIT), measured on Llama-3-8B against FP16, Wikitext-2, revision f364eb6f. Original table · Reading and choosing quantizations

GGUF source

DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF

Converted and published by a third party About this label

Last checked: Nov 11, 2026

Architecture: llama; Supported before llama.cpp b4409
Parameters: 405.9B
Max context length: 131,072
License: llama3.1 Conditional

File size and memory per quantization for shisa-v2-llama3.1-405b (DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF)
Quantization	File size Measured	Quality guide	4,096	8,192	16,384	32,768	65,536	131,072 max
Q2_K (11 parts)	149 GB	Very large	152 GB	154 GB	159 GB	167 GB	184 GB	218 GB
Q3_K_S (13 parts)	175 GB	Large	178 GB	180 GB	184 GB	193 GB	210 GB	244 GB
Q4_K_M (18 parts)	243 GB	Small	246 GB	248 GB	252 GB	261 GB	278 GB	312 GB
Q5_K_M (22 parts)	287 GB	Very small	290 GB	292 GB	296 GB	304 GB	321 GB	355 GB
Q6_K (25 parts)	333 GB	Very small	336 GB	338 GB	342 GB	351 GB	368 GB	401 GB
Q8_0 (32 parts)	431 GB	Nearly identical	434 GB	436 GB	440 GB	449 GB	466 GB	500 GB

Ollama  ollama run hf.co/DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF:Q4_K_M

llama.cpp  llama-server -hf DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF:Q4_K_M

GGUF source

shisa-ai/shisa-v2-llama3.1-405b-GGUF

Published by the original developer About this label

Last checked: Nov 11, 2026

Architecture: llama; Supported before llama.cpp b4409
Parameters: 405.9B
Max context length: 131,072
License: llama3.1 Conditional

File size and memory per quantization for shisa-v2-llama3.1-405b (shisa-ai/shisa-v2-llama3.1-405b-GGUF)
Quantization	File size Measured	Quality guide	4,096	8,192	16,384	32,768	65,536	131,072 max
IQ2_XXS (3 parts)	107 GB	Very large	110 GB	112 GB	117 GB	125 GB	142 GB	176 GB
IQ3_XS (4 parts)	166 GB	Large	169 GB	171 GB	175 GB	184 GB	201 GB	235 GB
IQ3_M (5 parts)	182 GB	Medium	185 GB	187 GB	191 GB	199 GB	216 GB	250 GB
IQ4_XS (5 parts)	217 GB	Small	219 GB	222 GB	226 GB	234 GB	251 GB	285 GB
Q4_K_M (6 parts)	243 GB	Small	246 GB	248 GB	252 GB	261 GB	278 GB	312 GB

Ollama  ollama run hf.co/shisa-ai/shisa-v2-llama3.1-405b-GGUF:Q4_K_M

llama.cpp  llama-server -hf shisa-ai/shisa-v2-llama3.1-405b-GGUF:Q4_K_M

Generation speed on common GPUs and Macs (estimated)

A rough guide to how fast Q4_K_M (243 GB) generates text. Each token reads the model weights once, so the upper limit is memory bandwidth ÷ file size. The guide is 60–80% of that, where most published measurements fall.

Device	Memory bandwidth	Guide (tokens/s) Estimated	Upper limit
Apple M5 Ultra 512 GB	1,200 GB/s	3.0–3.9	4.9

Estimates only. Prompt processing speed (limited by compute) is not included. Longer contexts and offloading part of the model to the CPU make it slower. Macs are shown only when installed memory is at least 4/3 of what is needed. Devices without published bandwidth (some GeForce cards, etc.) are not shown.

License and commercial use

Llama 3.1 Community License Conditional

Allows commercial use. Over 700 million monthly active users requires a separate license from Meta. Redistribution and published derivatives must show "Built with Llama", and you must follow the Acceptable Use Policy.

Based on the license on the original model's Hugging Face page. Original license text (checked Oct 18, 2026)

The summary states facts this site read from the original text and is not legal advice. Always check the original license and its latest terms before use. All licenses

How to run

Example commands using Q4_K_M. To use another quantization, replace the name with one from the table above. Commands for files from other sources are in each source's section. How to run local LLMs

Ollama

Run it by pointing directly at the GGUF on Hugging Face.

ollama run hf.co/DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF:Q4_K_M

llama.cpp

Pass the repository and quantization name to llama-server (or llama-cli) with -hf.

llama-server -hf DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF:Q4_K_M

LM Studio

Paste DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF into the search box, download it, and choose the quantization from the file list.

Frequently asked questions

How much memory do I need to run shisa-v2-llama3.1-405b (Q4_K_M)?: The file size is 243 GB (measured). With a 8,192-token context, the estimated memory including the KV cache is 248 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.
Can shisa-v2-llama3.1-405b be used commercially?: The original model's license is Llama 3.1 Community License (Commercial use with conditions).Allows commercial use. Over 700 million monthly active users requires a separate license from Meta. Redistribution and published derivatives must show "Built with Llama", and you must follow the Acceptable Use Policy. Check the original license text before deciding.
Does shisa-v2-llama3.1-405b run in 8 GB, 12 GB, 16 GB or 24 GB of VRAM?: 8 GB: no quantization fits; 12 GB: no quantization fits; 16 GB: no quantization fits; 24 GB: no quantization fits (estimated at 8,192 tokens with an f16 KV cache, files from the most-downloaded source).
What is the command to run shisa-v2-llama3.1-405b with Ollama?: Run "ollama run hf.co/DevQuasar/shisa-ai.shisa-v2-llama3.1-405b-GGUF:Q4_K_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.