How much memory do I need to run granite-4.2-30b (Q4_K_M)?

The file size is 17.7 GB (measured). With a 8,192-token context, the estimated memory including the KV cache is 20.6 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.

Does granite-4.2-30b 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: up to Q3_K_M (est. 17.0 GB); 24 GB: up to Q5_1 (est. 24.9 GB) (estimated at 8,192 tokens with an f16 KV cache, files from the most-downloaded source).

What is the command to run granite-4.2-30b with Ollama?

Run "ollama run hf.co/ibm-granite/granite-4.2-30b-GGUF:Q4_K_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.

ibm-granite

granite-4.2-30b

Parameters: 29.3B
Max context length: 131,072
Architecture: granite
License: Commercial OK

Base model: ibm-granite/granite-4.2-30b
License: apache-2.0

Memory needed to run Q4_K_M at 8K tokens

20.6 GB Estimated

File size 17.7 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 fits
32 GB fits
48 GB fits

See devices and models for 24 GB

The smallest, Q2_K, needs an estimated 13.7 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 granite-4.2-30b

granite-4.2-30b is a GGUF model based on ibm-granite/granite-4.2-30b on Hugging Face. GGUF files are published by ibm-granite/granite-4.2-30b-GGUF. It has about 29.3B parameters, the granite architecture, a maximum context length of 131,072 tokens (values written in the GGUF).

There are 15 quantizations, with file sizes from 10.9 GB (Q2_K) to 58.6 GB (BF16) (actual sizes on Hugging Face).

Running Q4_K_M (17.7 GB) with an 8,192-token context needs an estimated 20.6 GB of memory, and an estimated 27.1 GB at 32,768 tokens. For 8,192 tokens, a GPU with 24 GB or more of VRAM (such as GeForce RTX 3090 Ti, GeForce RTX 4090), or a Mac with 24 GB or more of unified memory, is a rough guide (macOS caps how much memory the GPU can use, so leave some headroom). The smallest, Q2_K (10.9 GB), needs an estimated 13.7 GB at 8,192 tokens.

The original model is licensed under Apache License 2.0, which allows commercial use (keep the license text and copyright notice when redistributing). 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).

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

ibm-granite/granite-4.2-30b-GGUF

Published by the original developer About this label

Last checked: Nov 11, 2026

Architecture: granite; Supported before llama.cpp b4409
Parameters: 29.3B
Max context length: 131,072
License: apache-2.0 Commercial OK

File size and memory per quantization for granite-4.2-30b (ibm-granite/granite-4.2-30b-GGUF)
Quantization	File size Measured	Quality guide	4,096	8,192	16,384	32,768	65,536	131,072 max
Q2_K	10.9 GB	Very large	12.7 GB	13.7 GB	15.9 GB	20.2 GB	28.8 GB	46.0 GB
Q3_K_S	12.7 GB	Large	14.6 GB	15.6 GB	17.8 GB	22.1 GB	30.7 GB	47.8 GB
Q3_K_M	14.1 GB	Medium	15.9 GB	17.0 GB	19.2 GB	23.5 GB	32.1 GB	49.2 GB
Q3_K_L	15.3 GB	Medium	17.1 GB	18.2 GB	20.3 GB	24.6 GB	33.2 GB	50.4 GB
Q4_0	16.6 GB	Medium	18.4 GB	19.5 GB	21.6 GB	25.9 GB	34.5 GB	51.7 GB
Q4_K_S	16.7 GB	Medium	18.5 GB	19.6 GB	21.8 GB	26.0 GB	34.6 GB	51.8 GB
Q4_K_M	17.7 GB	Small	19.5 GB	20.6 GB	22.8 GB	27.1 GB	35.6 GB	52.8 GB
Q4_1	18.4 GB	Medium	20.2 GB	21.3 GB	23.4 GB	27.7 GB	36.3 GB	53.5 GB
Q5_0	20.2 GB	Small	22.0 GB	23.1 GB	25.2 GB	29.5 GB	38.1 GB	55.3 GB
Q5_K_S	20.2 GB	Small	22.0 GB	23.1 GB	25.2 GB	29.5 GB	38.1 GB	55.3 GB
Q5_K_M	20.8 GB	Very small	22.6 GB	23.7 GB	25.8 GB	30.1 GB	38.7 GB	55.9 GB
Q5_1	22.0 GB	Small	23.8 GB	24.9 GB	27.0 GB	31.3 GB	39.9 GB	57.1 GB
Q6_K	24.0 GB	Very small	25.8 GB	26.9 GB	29.1 GB	33.4 GB	41.9 GB	59.1 GB
Q8_0	31.1 GB	Nearly identical	32.9 GB	34.0 GB	36.1 GB	40.4 GB	49.0 GB	66.2 GB
BF16 (2 parts)	58.6 GB	Original precision	60.4 GB	61.4 GB	63.6 GB	67.9 GB	76.5 GB	93.7 GB

Ollama  ollama run hf.co/ibm-granite/granite-4.2-30b-GGUF:Q4_K_M

llama.cpp  llama-server -hf ibm-granite/granite-4.2-30b-GGUF:Q4_K_M

Generation speed on common GPUs and Macs (estimated)

A rough guide to how fast Q4_K_M (17.7 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
Radeon RX 7900 XT 20 GB	800 GB/s	27–36	45
Radeon RX 7900 XTX 24 GB	960 GB/s	33–43	54
GeForce RTX 5090 32 GB	1,792 GB/s	61–81	101
Apple M4 32 GB	120 GB/s	4.1–5.4	6.8
Apple M4 Pro 48 GB	273 GB/s	9.2–12	15
Apple M5 Ultra 96 GB	1,200 GB/s	41–54	68

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.

Third-party evaluations

Values from other organizations for the original model before quantization. Not this site's own rating. Values are for the original model before quantization. GGUF quantization (especially small quantizations) can lower performance.

Human preference (LMArena)

1,340 1,329–1,350

Granite 4.2 30B · 3,314 votes · Values as of Nov 9, 2026

Source: LMArena (CC BY 4.0). Original leaderboard · See in ranking

License and commercial use

Apache License 2.0 Commercial OK

Allows commercial use, modification and redistribution. When redistributing, keep the license text and copyright notice, and note changes in modified files.

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. How to run local LLMs

Ollama

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

ollama run hf.co/ibm-granite/granite-4.2-30b-GGUF:Q4_K_M

llama.cpp

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

llama-server -hf ibm-granite/granite-4.2-30b-GGUF:Q4_K_M

LM Studio

Paste ibm-granite/granite-4.2-30b-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 granite-4.2-30b (Q4_K_M)?: The file size is 17.7 GB (measured). With a 8,192-token context, the estimated memory including the KV cache is 20.6 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.
Can granite-4.2-30b be used commercially?: The original model's license is Apache License 2.0 (Commercial use allowed).Allows commercial use, modification and redistribution. When redistributing, keep the license text and copyright notice, and note changes in modified files. Check the original license text before deciding.
Does granite-4.2-30b 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: up to Q3_K_M (est. 17.0 GB); 24 GB: up to Q5_1 (est. 24.9 GB) (estimated at 8,192 tokens with an f16 KV cache, files from the most-downloaded source).
What is the command to run granite-4.2-30b with Ollama?: Run "ollama run hf.co/ibm-granite/granite-4.2-30b-GGUF:Q4_K_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.