Qwen3.8 27B vs MiMo V2.5

Qwen3.8 27B leads llmrun Score 121.5 to 108.3; only Qwen3.8 27B fits 24 GB (Q4_K_M); MiMo V2.5 needs 86.6 GB even at its smallest quant.

Model A · Alibaba

Qwen3.8 27B

121.5llmrun Score

27.8B · 256K context · Apache 2.0 · Aug 2026

VisionFunctions

Model B · Xiaomi

MiMo V2.5

108.3llmrun Score

310.8B · 17.4B active · 1M context · MIT · Apr 2026

ChatFunctions

Which should you choose?

Choose Qwen3.8 27B if you want…

Higher llmrun Score
121.5 vs 108.3
Stronger at coding
112.1 vs 103.3
Fits a 12 GB GPU
IQ3_XXS · 12 GB; MiMo V2.5 needs 86.6 GB at its smallest quant
Fits a Mac 32 GB
Q4_K_M · 17.9 GB; MiMo V2.5 needs 86.6 GB at its smallest quant
Lighter on memory
17.4 vs 186.9 GB at Q4_K_M
Also does vision
MiMo V2.5 doesn't list it

Choose MiMo V2.5 if you want…

Longer context window
1M vs 256K tokens
Also does chat
Qwen3.8 27B doesn't list it

On your hardware

Each model at its default quant (Q4_K_M), a smaller one when it must. Estimates, ±20%.

MemoryQwen3.8 27BMiMo V2.5

8 GB

RTX 4060 · 8 GB tier

Qwen3.8 27B

Doesn't fit

smallest quant needs 8.8 GB

MiMo V2.5

Doesn't fit

smallest quant needs 86.6 GB

12 GB

RTX 3060 12GB · 12 GB tier

Qwen3.8 27B

IQ3_XXS · ~20 tok/s

12 GB · up to 16Kfits

MiMo V2.5

Doesn't fit

smallest quant needs 86.6 GB

16 GB

RTX 5060 Ti 16GB · 16 GB tier

Qwen3.8 27B

Q3_K_L · ~19 tok/s

15.4 GB · up to 16Kfits

MiMo V2.5

Doesn't fit

smallest quant needs 86.6 GB

24 GBreference

RTX 4090 · 24 GB tier

Qwen3.8 27B

Q4_K_M · ~38 tok/s

17.9 GB · up to 64Kfits

MiMo V2.5

Doesn't fit

smallest quant needs 86.6 GB

32 GB

RTX 5090 · 32 GB tier

Qwen3.8 27B

Q4_K_M · ~67 tok/s

17.9 GB · up to 256Kfits

MiMo V2.5

Doesn't fit

smallest quant needs 86.6 GB

48 GB

RTX A6000 · 48 GB tier

Qwen3.8 27B

Q4_K_M · ~29 tok/s

17.9 GB · up to 256Kfits

MiMo V2.5

Doesn't fit

smallest quant needs 86.6 GB

Mac 32 GB

Mac Mini M4 · Macs

Qwen3.8 27B

Q4_K_M · ~5 tok/s

17.9 GB · up to 128Kfits

MiMo V2.5

Doesn't fit

smallest quant needs 86.6 GB

Mac 64 GB

Mac Studio M4 Max · Macs

Qwen3.8 27B

Q4_K_M · ~22 tok/s

17.9 GB · up to 256Kfits

MiMo V2.5

Doesn't fit

smallest quant needs 86.6 GB

Mac 128 GB

Mac Studio M4 Max · Macs

Qwen3.8 27B

Q4_K_M · ~22 tok/s

17.9 GB · up to 256Kmore headroom

MiMo V2.5

IQ3_XXS · ~37 tok/s

121.5 GB · up to 64K

Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).

llmrun Score

with its 90% range · bars share one scale

Qwen3.8 27B MiMo V2.5

Overall

Qwen3.8 27B

121.5119–124

MiMo V2.5

108.399–118

Coding

Qwen3.8 27B

112.1105–119

MiMo V2.5

103.390–116

Math

Qwen3.8 27B

117.5111–125

MiMo V2.5

no data

Science

Qwen3.8 27B

112.093–131

MiMo V2.5

106.987–127

Reasoning

Qwen3.8 27B

124.1121–127

MiMo V2.5

no data

Qwen3.8 27B · #10 of 92 open models · #1 of 34 that fit 24 GB

MiMo V2.5 · #30 of 92 open models · #13 of 71 that fit ≤192 GB

Benchmarks

5 shared boards: Qwen3.8 27B leads 3, MiMo V2.5 leads 0 · Benchmarks imported Oct 2026

BoardQwen3.8 27BMiMo V2.5

LMArena Text

Qwen3.8 27B

1438.1#22/186 open

ArenaLMArena · Oct 2026

MiMo V2.5

1433.7#24/186 open

ArenaLMArena · Oct 2026

LMArena WebDev

Qwen3.8 27B

1592.8#10/50 open

ArenaLMArena · Oct 2026

MiMo V2.5

1438.1#20/50 open

ArenaLMArena · Oct 2026

SciCode

Qwen3.8 27B

46.6%#19/65 open

Lab-reportedEpoch AI · Aug 2026

MiMo V2.5

43.1%#24/65 open

Lab-reportedEpoch AI

CritPt

Qwen3.8 27B

5.4%#15/65 open

Lab-reportedEpoch AI · Aug 2026

MiMo V2.5

3.7%#21/65 open

Lab-reportedEpoch AI

ProofBench

Qwen3.8 27B

16.0%#16/26 open

Lab-reportedEpoch AI · Aug 2026

MiMo V2.5

16.0%#17/26 open

Lab-reportedEpoch AI

10 more boards (other shared and single-model boards)

Only Qwen3.8 27B (9)

Only MiMo V2.5 (1)

ALE-Bench514.0

VRAM by quantization

Memory to load the model and 2K context; below it, the total at 32K and 128K context. Lighter side in bold.

QuantQwen3.8 27BMiMo V2.5

Q2_K

Qwen3.8 27B

12.6 GB10.8 GB file

32K 13.9 GB · 128K 19.3 GB

MiMo V2.5

132 GB108.9 GB file

32K 134 GB · 128K 139 GB

Q3_K_S

Qwen3.8 27B

12.9 GB12.7 GB file

32K 14.2 GB · 128K 19.6 GB

MiMo V2.5

136 GB135.1 GB file

32K 138 GB · 128K 143 GB

Q3_K_M

Qwen3.8 27B

14.3 GB13.4 GB file

32K 15.6 GB · 128K 21.0 GB

MiMo V2.5

152 GB140.0 GB file

32K 153 GB · 128K 158 GB

Q4_0

Qwen3.8 27B

14.6 GB~13.9 GB file

32K 16.0 GB · 128K 21.4 GB

MiMo V2.5

156 GB175.8 GB file

32K 157 GB · 128K 162 GB

Q4_K_M

Qwen3.8 27B

17.4 GB16.5 GB file

32K 18.8 GB · 128K 24.1 GB

MiMo V2.5

187 GB190.9 GB file

32K 188 GB · 128K 193 GB

Q5_K_M

Qwen3.8 27B

20.5 GB19.8 GB file

32K 21.9 GB · 128K 27.3 GB

MiMo V2.5

222 GB230.2 GB file

32K 223 GB · 128K 228 GB

Q6_K

Qwen3.8 27B

23.7 GB22.0 GB file

32K 25.0 GB · 128K 30.4 GB

MiMo V2.5

257 GB257.0 GB file

32K 258 GB · 128K 263 GB

Q8_0

Qwen3.8 27B

28.5 GB29.0 GB file

32K 29.9 GB · 128K 35.2 GB

MiMo V2.5

311 GB329.3 GB file

32K 313 GB · 128K 318 GB

17 more quants

IQ2_XXS

Qwen3.8 27B

8.4 GB7.3 GB file

32K 9.7 GB · 128K 15.1 GB

MiMo V2.5

85.9 GB96.5 GB file

32K 87.4 GB · 128K 92.2 GB

IQ2_XS

Qwen3.8 27B

9.1 GB9.1 GB file

32K 10.4 GB · 128K 15.8 GB

MiMo V2.5

93.6 GB93.1 GB file

32K 95.1 GB · 128K 100 GB

IQ2_S

Qwen3.8 27B

9.4 GB8.4 GB file

32K 10.8 GB · 128K 16.1 GB

MiMo V2.5

97.5 GB94.7 GB file

32K 99.0 GB · 128K 104 GB

IQ2_M

Qwen3.8 27B

10.1 GB10.5 GB file

32K 11.5 GB · 128K 16.8 GB

MiMo V2.5

105 GB96.6 GB file

32K 107 GB · 128K 112 GB

IQ3_XXS

Qwen3.8 27B

11.5 GB10.9 GB file

32K 12.9 GB · 128K 18.2 GB

MiMo V2.5

121 GB125.9 GB file

32K 122 GB · 128K 127 GB

IQ3_XS

Qwen3.8 27B

12.2 GB10.4 GB file

32K 13.6 GB · 128K 18.9 GB

MiMo V2.5

129 GB142.2 GB file

32K 130 GB · 128K 135 GB

IQ3_S

Qwen3.8 27B

12.6 GB12.0 GB file

32K 13.9 GB · 128K 19.3 GB

MiMo V2.5

132 GB114.9 GB file

32K 134 GB · 128K 139 GB

IQ3_M

Qwen3.8 27B

13.3 GB14.9 GB file

32K 14.6 GB · 128K 20.0 GB

MiMo V2.5

140 GB148.3 GB file

32K 142 GB · 128K 147 GB

Q3_K_L

Qwen3.8 27B

15.0 GB14.1 GB file

32K 16.3 GB · 128K 21.7 GB

MiMo V2.5

160 GB147.9 GB file

32K 161 GB · 128K 166 GB

IQ4_XS

Qwen3.8 27B

15.7 GB14.3 GB file

32K 17.0 GB · 128K 22.4 GB

MiMo V2.5

167 GB149.4 GB file

32K 169 GB · 128K 174 GB

IQ4_NL

Qwen3.8 27B

16.4 GB17.4 GB file

32K 17.7 GB · 128K 23.1 GB

MiMo V2.5

175 GB152.6 GB file

32K 177 GB · 128K 182 GB

Q4_1

Qwen3.8 27B

16.4 GB17.5 GB file

32K 17.7 GB · 128K 23.1 GB

MiMo V2.5

175 GB194.3 GB file

32K 177 GB · 128K 182 GB

Q4_K_S

Qwen3.8 27B

16.4 GB15.4 GB file

32K 17.7 GB · 128K 23.1 GB

MiMo V2.5

175 GB178.8 GB file

32K 177 GB · 128K 182 GB

Q4_K_L

Qwen3.8 27B

17.8 GB18.8 GB file

32K 19.1 GB · 128K 24.5 GB

MiMo V2.5

191 GB189.2 GB file

32K 192 GB · 128K 197 GB

Q5_K_S

Qwen3.8 27B

19.9 GB18.7 GB file

32K 21.2 GB · 128K 26.6 GB

MiMo V2.5

214 GB216.3 GB file

32K 216 GB · 128K 220 GB

Q5_K_L

Qwen3.8 27B

20.9 GB21.5 GB file

32K 22.2 GB · 128K 27.6 GB

MiMo V2.5

not listed

BF16

Qwen3.8 27B

56.3 GB54.7 GB file

32K 57.7 GB · 128K 63.0 GB

MiMo V2.5

622 GB619.6 GB file

32K 623 GB · 128K 628 GB

How to run each

Qwen3.8 27B

Ollama · Q4_K_M

$ollama run hf.co/unsloth/Qwen3.8-27B-GGUF:Q4_K_M

All ways to run Qwen3.8 27B →

MiMo V2.5

Ollama · Q4_K_M

$ollama run hf.co/unsloth/MiMo-V2.5-GGUF:Q4_K_M

All ways to run MiMo V2.5 →

Specifications

Specifications of Qwen3.8 27B and MiMo V2.5
Spec	Qwen3.8 27B	MiMo V2.5
Publisher	Alibaba	Xiaomi
Parameters	27.8B	310.8B
Active parameters	—	17B per token
Architecture	Qwen3_5ForConditionalGeneration	MiMoV2ForCausalLM
Context	256K tokens	1M tokens
Modalities	Vision, Function-calling	Chat, Function-calling
License	Apache 2.0	MIT
Released	Aug 2026	Apr 2026

Frequently Asked Questions

Which is better, Qwen3.8 27B or MiMo V2.5?

Qwen3.8 27B leads llmrun Score 121.5 to 108.3; only Qwen3.8 27B fits 24 GB (Q4_K_M); MiMo V2.5 needs 86.6 GB even at its smallest quant. Choose Qwen3.8 27B for higher llmrun Score and stronger at coding; choose MiMo V2.5 for longer context window and also does chat.

Which needs less VRAM, Qwen3.8 27B or MiMo V2.5?

At Q4_K_M, Qwen3.8 27B requires 17.4 GB while MiMo V2.5 needs 186.9 GB, making Qwen3.8 27B the easier model to run on your local setup.

Can I run Qwen3.8 27B or MiMo V2.5 on a 16 GB GPU?

Qwen3.8 27B fits at Q3_K_L; MiMo V2.5 does not fit (smallest quant needs 86.6 GB).

Which has a longer context window, Qwen3.8 27B or MiMo V2.5?

Qwen3.8 27B supports 262,144 tokens and MiMo V2.5 supports 1,048,576 tokens.