How much memory do I need to run MiMo-V2-Flash (Q4_K_M)?

The file size is 187 GB (measured). With a 8,192-token context, the estimated memory including the KV cache is 188 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.

Does MiMo-V2-Flash 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 MiMo-V2-Flash with Ollama?

Run "ollama run hf.co/unsloth/MiMo-V2-Flash-GGUF:Q4_K_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.

XiaomiMiMo

MiMo-V2-Flash

Parameters: 308.8B
Max context length: 262,144
Architecture: mimo2
License: Commercial OK

Base model: XiaomiMiMo/MiMo-V2-Flash
License: mit
llama.cpp: b7536+

Memory needed to run Q4_K_M at 8K tokens

188 GB Estimated

File size 187 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

See devices and models for 192 GB

The smallest, UD-TQ1_0, needs an estimated 79.1 GB at 8K.

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

Data as of Oct 31, 2026

About MiMo-V2-Flash

MiMo-V2-Flash is a GGUF model based on XiaomiMiMo/MiMo-V2-Flash on Hugging Face. GGUF files are published by unsloth/MiMo-V2-Flash-GGUF. It has about 308.8B parameters, the mimo2 architecture, a maximum context length of 262,144 tokens (values written in the GGUF).

There are 27 quantizations, with file sizes from 77.7 GB (UD-TQ1_0) to 618 GB (BF16) (actual sizes on Hugging Face).

Running Q4_K_M (187 GB) with an 8,192-token context needs an estimated 188 GB of memory, and an estimated 189 GB at 32,768 tokens. For 8,192 tokens, a GPU with 192 GB or more of VRAM, or a Mac with 192 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, UD-TQ1_0 (77.7 GB), needs an estimated 79.1 GB at 8,192 tokens.

The original model is licensed under MIT License, 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

unsloth/MiMo-V2-Flash-GGUF

Converted and published by a third party About this label

Last checked: Oct 31, 2026

Architecture: mimo2; llama.cpp b7536 and later
Parameters: 308.8B
Max context length: 262,144
License: mit Commercial OK

File size and memory per quantization for MiMo-V2-Flash (unsloth/MiMo-V2-Flash-GGUF)
Quantization	File size Measured	Quality guide	4,096	8,192	16,384	32,768	65,536	131,072	262,144 max
UD-TQ1_0	77.7 GB	-	79.0 GB	79.1 GB	79.3 GB	79.7 GB	80.4 GB	81.9 GB	85.0 GB
UD-IQ1_S (2 parts)	87.2 GB	-	88.5 GB	88.6 GB	88.8 GB	89.2 GB	90.0 GB	91.5 GB	94.5 GB
UD-IQ1_M (2 parts)	93.8 GB	-	95.1 GB	95.2 GB	95.4 GB	95.8 GB	96.5 GB	98.0 GB	101 GB
UD-IQ2_XXS (3 parts)	102 GB	-	103 GB	103 GB	103 GB	104 GB	104 GB	106 GB	109 GB
UD-IQ2_M (3 parts)	107 GB	-	108 GB	108 GB	108 GB	109 GB	109 GB	111 GB	114 GB
Q2_K (3 parts)	112 GB	Very large	114 GB	114 GB	114 GB	114 GB	115 GB	117 GB	120 GB
Q2_K_L (3 parts)	112 GB	-	114 GB	114 GB	114 GB	114 GB	115 GB	117 GB	120 GB
UD-Q2_K_XL (3 parts)	116 GB	-	118 GB	118 GB	118 GB	118 GB	119 GB	121 GB	124 GB
UD-IQ3_XXS (3 parts)	127 GB	-	129 GB	129 GB	129 GB	129 GB	130 GB	132 GB	135 GB
Q3_K_S (3 parts)	133 GB	Large	134 GB	135 GB	135 GB	135 GB	136 GB	137 GB	140 GB
UD-Q3_K_XL (3 parts)	137 GB	-	138 GB	138 GB	139 GB	139 GB	140 GB	141 GB	144 GB
Q3_K_M (3 parts)	147 GB	Medium	149 GB	149 GB	149 GB	149 GB	150 GB	152 GB	155 GB
IQ4_XS (4 parts)	164 GB	Small	166 GB	166 GB	166 GB	166 GB	167 GB	169 GB	172 GB
IQ4_NL (4 parts)	174 GB	Small	175 GB	175 GB	176 GB	176 GB	177 GB	178 GB	181 GB
Q4_0 (4 parts)	175 GB	Medium	176 GB	176 GB	176 GB	177 GB	177 GB	179 GB	182 GB
Q4_K_S (4 parts)	175 GB	Medium	177 GB	177 GB	177 GB	177 GB	178 GB	180 GB	183 GB
UD-Q4_K_XL (4 parts)	178 GB	-	179 GB	179 GB	179 GB	180 GB	180 GB	182 GB	185 GB
Q4_K_M (4 parts)	187 GB	Small	188 GB	188 GB	188 GB	189 GB	190 GB	191 GB	194 GB
Q4_1 (4 parts)	193 GB	Medium	195 GB	195 GB	195 GB	195 GB	196 GB	198 GB	201 GB
Q5_K_S (5 parts)	213 GB	Small	214 GB	214 GB	214 GB	215 GB	215 GB	217 GB	220 GB
Q5_K_M (5 parts)	219 GB	Very small	220 GB	221 GB	221 GB	221 GB	222 GB	223 GB	226 GB
UD-Q5_K_XL (5 parts)	220 GB	-	221 GB	221 GB	221 GB	222 GB	223 GB	224 GB	227 GB
Q6_K (6 parts)	253 GB	Very small	255 GB	255 GB	255 GB	255 GB	256 GB	258 GB	261 GB
UD-Q6_K_XL (6 parts)	263 GB	-	265 GB	265 GB	265 GB	265 GB	266 GB	268 GB	271 GB
Q8_0 (7 parts)	328 GB	Nearly identical	330 GB	330 GB	330 GB	330 GB	331 GB	332 GB	335 GB
UD-Q8_K_XL (8 parts)	354 GB	-	356 GB	356 GB	356 GB	356 GB	357 GB	358 GB	361 GB
BF16 (13 parts)	618 GB	Original precision	619 GB	619 GB	619 GB	620 GB	620 GB	622 GB	625 GB

Ollama  ollama run hf.co/unsloth/MiMo-V2-Flash-GGUF:Q4_K_M

llama.cpp  llama-server -hf unsloth/MiMo-V2-Flash-GGUF:Q4_K_M

License and commercial use

MIT License Commercial OK

Allows commercial use, modification and redistribution. When redistributing, keep the copyright notice and license text.

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

Mainline llama.cpp can load this model's format (mimo2) from b7536 (Dec 24, 2025) onward. Ollama and LM Studio bundle llama.cpp internally, so an older app fails to load it with errors such as "unknown model architecture". If that happens, update the app.

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/unsloth/MiMo-V2-Flash-GGUF:Q4_K_M

llama.cpp

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

llama-server -hf unsloth/MiMo-V2-Flash-GGUF:Q4_K_M

LM Studio

Paste unsloth/MiMo-V2-Flash-GGUF into the search box, download it, and choose the quantization from the file list.

XiaomiMiMo

MiMo-V2.6-Distill-Qwen-9B

9B

Q4_K_M size

5.84 GB

Memory at 8K

7.21 GB

Measured Estimated VRAM 8 GB+

Frequently asked questions

How much memory do I need to run MiMo-V2-Flash (Q4_K_M)?: The file size is 187 GB (measured). With a 8,192-token context, the estimated memory including the KV cache is 188 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.
Can MiMo-V2-Flash be used commercially?: The original model's license is MIT License (Commercial use allowed).Allows commercial use, modification and redistribution. When redistributing, keep the copyright notice and license text. Check the original license text before deciding.
Does MiMo-V2-Flash 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 MiMo-V2-Flash with Ollama?: Run "ollama run hf.co/unsloth/MiMo-V2-Flash-GGUF:Q4_K_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.