How much memory do I need to run Swallow-7b-instruct-hf (Q4_K_M)?

The file size is 4.14 GB (measured). With a 4,096-token context, the estimated memory including the KV cache is 6.92 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.

What is the command to run Swallow-7b-instruct-hf with Ollama?

Run "ollama run hf.co/mradermacher/Swallow-7b-instruct-hf-GGUF:Q4_K_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.

tokyotech-llm

Swallow-7b-instruct-hf

Parameters: 6.8B
Max context length: 4,096
Architecture: llama
License: Conditional

Base model: tokyotech-llm/Swallow-7b-instruct-hf
License: llama2

About Swallow-7b-instruct-hf

Swallow-7b-instruct-hf is a GGUF model based on tokyotech-llm/Swallow-7b-instruct-hf on Hugging Face. GGUF files are published by 2 sources (mradermacher/Swallow-7b-instruct-hf-GGUF, mradermacher/Swallow-7b-instruct-hf-i1-GGUF). It has about 6.8B parameters, the llama architecture, a maximum context length of 4,096 tokens (values written in the GGUF).

There are 23 quantizations, with file sizes from 1.58 GB (IQ1_S) to 13.7 GB (F16) (actual sizes on Hugging Face).

The original model is licensed under Llama 2 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

mradermacher/Swallow-7b-instruct-hf-GGUF

Converted and published by a third party About this label

Last checked: Nov 11, 2026

Architecture: llama; Supported before llama.cpp b4409
Parameters: 6.8B
Max context length: 4,096
License: llama2 Conditional

File size and memory per quantization for Swallow-7b-instruct-hf (mradermacher/Swallow-7b-instruct-hf-GGUF)
Quantization	File size Measured	Quality guide	4,096 max
Q2_K	2.59 GB	Very large	5.36 GB
IQ3_XS	2.85 GB	Large	5.63 GB
IQ3_S	3.01 GB	Medium	5.78 GB
Q3_K_S	3.01 GB	Large	5.78 GB
IQ3_M	3.17 GB	Medium	5.95 GB
Q3_K_M	3.36 GB	Medium	6.13 GB
Q3_K_L	3.65 GB	Medium	6.43 GB
IQ4_XS	3.71 GB	Small	6.48 GB
Q4_K_S	3.92 GB	Medium	6.69 GB
Q4_K_M	4.14 GB	Small	6.92 GB
Q5_K_S	4.72 GB	Small	7.49 GB
Q5_K_M	4.85 GB	Very small	7.63 GB
Q6_K	5.60 GB	Very small	8.38 GB
Q8_0	7.26 GB	Nearly identical	10.0 GB
F16	13.7 GB	Original precision	16.4 GB

Ollama  ollama run hf.co/mradermacher/Swallow-7b-instruct-hf-GGUF:Q4_K_M

llama.cpp  llama-server -hf mradermacher/Swallow-7b-instruct-hf-GGUF:Q4_K_M

GGUF source

mradermacher/Swallow-7b-instruct-hf-i1-GGUF

Converted and published by a third party About this label

Last checked: Nov 11, 2026

Architecture: llama; Supported before llama.cpp b4409
Parameters: 6.8B
Max context length: 4,096
License: llama2 Conditional

File size and memory per quantization for Swallow-7b-instruct-hf (mradermacher/Swallow-7b-instruct-hf-i1-GGUF)
Quantization	File size Measured	Quality guide	4,096 max
IQ1_S	1.58 GB	Very large	4.35 GB
IQ1_M	1.70 GB	Very large	4.47 GB
IQ2_XXS	1.90 GB	Very large	4.67 GB
IQ2_XS	2.08 GB	Very large	4.85 GB
IQ2_S	2.25 GB	Very large	5.02 GB
IQ2_M	2.41 GB	Large	5.18 GB
Q2_K	2.59 GB	Very large	5.36 GB
IQ3_XXS	2.64 GB	Large	5.41 GB
IQ3_XS	2.85 GB	Large	5.63 GB
Q3_K_S	3.01 GB	Large	5.78 GB
IQ3_S	3.01 GB	Medium	5.78 GB
IQ3_M	3.17 GB	Medium	5.95 GB
Q3_K_M	3.36 GB	Medium	6.13 GB
Q3_K_L	3.65 GB	Medium	6.43 GB
IQ4_XS	3.68 GB	Small	6.45 GB
Q4_0	3.90 GB	Medium	6.67 GB
Q4_K_S	3.92 GB	Medium	6.69 GB
Q4_K_M	4.14 GB	Small	6.92 GB
Q5_K_S	4.72 GB	Small	7.49 GB
Q5_K_M	4.85 GB	Very small	7.63 GB
Q6_K	5.60 GB	Very small	8.38 GB

Ollama  ollama run hf.co/mradermacher/Swallow-7b-instruct-hf-i1-GGUF:Q4_K_M

llama.cpp  llama-server -hf mradermacher/Swallow-7b-instruct-hf-i1-GGUF:Q4_K_M

License and commercial use

Llama 2 Community License Conditional

Allows commercial use. However, products or services with more than 700 million monthly active users need a separate license from Meta. 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/mradermacher/Swallow-7b-instruct-hf-GGUF:Q4_K_M

llama.cpp

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

llama-server -hf mradermacher/Swallow-7b-instruct-hf-GGUF:Q4_K_M

LM Studio

Paste mradermacher/Swallow-7b-instruct-hf-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 Swallow-7b-instruct-hf (Q4_K_M)?: The file size is 4.14 GB (measured). With a 4,096-token context, the estimated memory including the KV cache is 6.92 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.
Can Swallow-7b-instruct-hf be used commercially?: The original model's license is Llama 2 Community License (Commercial use with conditions).Allows commercial use. However, products or services with more than 700 million monthly active users need a separate license from Meta. You must follow the Acceptable Use Policy. Check the original license text before deciding.
What is the command to run Swallow-7b-instruct-hf with Ollama?: Run "ollama run hf.co/mradermacher/Swallow-7b-instruct-hf-GGUF:Q4_K_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.