Qwen3 235B A22B Instruct 2507 vs Qwen3.8 27B

Qwen3.8 27B leads llmrun Score 121.5 to 98.5; only Qwen3.8 27B fits 24 GB (Q4_K_M); Qwen3 235B A22B Instruct 2507 needs 72.4 GB even at its smallest quant.

Model A · Alibaba

Qwen3 235B A22B Instruct 2507

98.5llmrun Score

235.1B · 22.2B active · 256K context · Apache 2.0 · Jul 2025

ChatFunctions

Model B · Alibaba

Qwen3.8 27B

121.5llmrun Score

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

VisionFunctions

Which should you choose?

Choose Qwen3 235B A22B Instruct 2507 if you want…

Also does chat
Qwen3.8 27B doesn't list it

Choose Qwen3.8 27B if you want…

Higher llmrun Score
121.5 vs 98.5
Stronger at reasoning
124.1 vs 98.1
Stronger at coding
112.1 vs 101.6
Ahead on the benchmarks
5 of 5 shared boards
Fits a 12 GB GPU
IQ3_XXS · 12 GB; Qwen3 235B A22B Instruct 2507 needs 72.4 GB at its smallest quant
Fits a Mac 32 GB
Q4_K_M · 17.9 GB; Qwen3 235B A22B Instruct 2507 needs 72.4 GB at its smallest quant

On your hardware

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

MemoryQwen3 235B A22B Instruct 2507Qwen3.8 27B

8 GB

RTX 4060 · 8 GB tier

Qwen3 235B A22B Instruct 2507

Doesn't fit

smallest quant needs 72.4 GB

Qwen3.8 27B

Doesn't fit

smallest quant needs 8.8 GB

12 GB

RTX 3060 12GB · 12 GB tier

Qwen3 235B A22B Instruct 2507

Doesn't fit

smallest quant needs 72.4 GB

Qwen3.8 27B

IQ3_XXS · ~20 tok/s

12 GB · up to 16Kfits

16 GB

RTX 5060 Ti 16GB · 16 GB tier

Qwen3 235B A22B Instruct 2507

Doesn't fit

smallest quant needs 72.4 GB

Qwen3.8 27B

Q3_K_L · ~19 tok/s

15.4 GB · up to 16Kfits

24 GBreference

RTX 4090 · 24 GB tier

Qwen3 235B A22B Instruct 2507

Doesn't fit

smallest quant needs 72.4 GB

Qwen3.8 27B

Q4_K_M · ~38 tok/s

17.9 GB · up to 64Kfits

32 GB

RTX 5090 · 32 GB tier

Qwen3 235B A22B Instruct 2507

Doesn't fit

smallest quant needs 72.4 GB

Qwen3.8 27B

Q4_K_M · ~67 tok/s

17.9 GB · up to 256Kfits

48 GB

RTX A6000 · 48 GB tier

Qwen3 235B A22B Instruct 2507

Doesn't fit

smallest quant needs 72.4 GB

Qwen3.8 27B

Q4_K_M · ~29 tok/s

17.9 GB · up to 256Kfits

Mac 32 GB

Mac Mini M4 · Macs

Qwen3 235B A22B Instruct 2507

Doesn't fit

smallest quant needs 72.4 GB

Qwen3.8 27B

Q4_K_M · ~5 tok/s

17.9 GB · up to 128Kfits

Mac 64 GB

Mac Studio M4 Max · Macs

Qwen3 235B A22B Instruct 2507

Doesn't fit

smallest quant needs 72.4 GB

Qwen3.8 27B

Q4_K_M · ~22 tok/s

17.9 GB · up to 256Kfits

Mac 128 GB

Mac Studio M4 Max · Macs

Qwen3 235B A22B Instruct 2507

Q3_K_L · ~21 tok/s

122.4 GB · up to 32K

Qwen3.8 27B

Q4_K_M · ~22 tok/s

17.9 GB · up to 256Kmore headroom

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 235B A22B Instruct 2507 Qwen3.8 27B

Overall

Qwen3 235B A22B Instruct 2507

98.595–102

Qwen3.8 27B

121.5119–124

Coding

Qwen3 235B A22B Instruct 2507

101.696–107

Qwen3.8 27B

112.1105–119

Math

Qwen3 235B A22B Instruct 2507

no data

Qwen3.8 27B

117.5111–125

Science

Qwen3 235B A22B Instruct 2507

no data

Qwen3.8 27B

112.093–131

Reasoning

Qwen3 235B A22B Instruct 2507

98.193–104

Qwen3.8 27B

124.1121–127

Qwen3 235B A22B Instruct 2507 · #51 of 92 open models · #33 of 71 that fit ≤192 GB

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

Benchmarks

5 shared boards: Qwen3 235B A22B Instruct 2507 leads 0, Qwen3.8 27B leads 5 · Benchmarks imported Oct 2026

BoardQwen3 235B A22B Instruct 2507Qwen3.8 27B

LMArena Text

Qwen3 235B A22B Instruct 2507

1422.3#32/186 open

ArenaLMArena · Oct 2026

Qwen3.8 27B

1438.1#22/186 open

ArenaLMArena · Oct 2026

ARC-AGI-2

Qwen3 235B A22B Instruct 2507

1.3%#14/18 open

IndependentEpoch AI · Jul 2025

Qwen3.8 27B

42.4%#5/18 open

IndependentEpoch AI · Aug 2026

ARC-AGI

Qwen3 235B A22B Instruct 2507

11.0%#15/18 open

IndependentEpoch AI · Jul 2025

Qwen3.8 27B

87.5%#5/18 open

IndependentEpoch AI · Aug 2026

DTBench

Qwen3 235B A22B Instruct 2507

78.4%#25/68 open

Lab-reportedEpoch AI · Jul 2025

Qwen3.8 27B

88.0%#9/68 open

Lab-reportedEpoch AI · Aug 2026

LMCA

Qwen3 235B A22B Instruct 2507

23.5%#30/54 open

Lab-reportedEpoch AI · Jul 2025

Qwen3.8 27B

41.4%#7/54 open

Lab-reportedEpoch AI · Aug 2026

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

Only Qwen3 235B A22B Instruct 2507 (4)

Only Qwen3.8 27B (9)

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 235B A22B Instruct 2507Qwen3.8 27B

Q2_K

Qwen3 235B A22B Instruct 2507

100 GB85.7 GB file

32K 103 GB · 128K 113 GB

Qwen3.8 27B

12.6 GB10.8 GB file

32K 13.9 GB · 128K 19.3 GB

Q3_K_S

Qwen3 235B A22B Instruct 2507

103 GB101.4 GB file

32K 106 GB · 128K 116 GB

Qwen3.8 27B

12.9 GB12.7 GB file

32K 14.2 GB · 128K 19.6 GB

Q3_K_M

Qwen3 235B A22B Instruct 2507

115 GB112.4 GB file

32K 118 GB · 128K 128 GB

Qwen3.8 27B

14.3 GB13.4 GB file

32K 15.6 GB · 128K 21.0 GB

Q4_0

Qwen3 235B A22B Instruct 2507

118 GB133.1 GB file

32K 121 GB · 128K 130 GB

Qwen3.8 27B

14.6 GB~13.9 GB file

32K 16.0 GB · 128K 21.4 GB

Q4_K_M

Qwen3 235B A22B Instruct 2507

142 GB142.2 GB file

32K 145 GB · 128K 154 GB

Qwen3.8 27B

17.4 GB16.5 GB file

32K 18.8 GB · 128K 24.1 GB

Q5_K_M

Qwen3 235B A22B Instruct 2507

168 GB166.8 GB file

32K 171 GB · 128K 180 GB

Qwen3.8 27B

20.5 GB19.8 GB file

32K 21.9 GB · 128K 27.3 GB

Q6_K

Qwen3 235B A22B Instruct 2507

194 GB193.0 GB file

32K 197 GB · 128K 207 GB

Qwen3.8 27B

23.7 GB22.0 GB file

32K 25.0 GB · 128K 30.4 GB

Q8_0

Qwen3 235B A22B Instruct 2507

236 GB249.9 GB file

32K 239 GB · 128K 248 GB

Qwen3.8 27B

28.5 GB29.0 GB file

32K 29.9 GB · 128K 35.2 GB

17 more quants

IQ2_XXS

Qwen3 235B A22B Instruct 2507

not listed

Qwen3.8 27B

8.4 GB7.3 GB file

32K 9.7 GB · 128K 15.1 GB

IQ2_XS

Qwen3 235B A22B Instruct 2507

71.0 GB65.6 GB file

32K 74.0 GB · 128K 83.5 GB

Qwen3.8 27B

9.1 GB9.1 GB file

32K 10.4 GB · 128K 15.8 GB

IQ2_S

Qwen3 235B A22B Instruct 2507

74.0 GB66.1 GB file

32K 76.9 GB · 128K 86.4 GB

Qwen3.8 27B

9.4 GB8.4 GB file

32K 10.8 GB · 128K 16.1 GB

IQ2_M

Qwen3 235B A22B Instruct 2507

79.8 GB78.6 GB file

32K 82.8 GB · 128K 92.3 GB

Qwen3.8 27B

10.1 GB10.5 GB file

32K 11.5 GB · 128K 16.8 GB

IQ3_XXS

Qwen3 235B A22B Instruct 2507

91.6 GB93.1 GB file

32K 94.6 GB · 128K 104 GB

Qwen3.8 27B

11.5 GB10.9 GB file

32K 12.9 GB · 128K 18.2 GB

IQ3_XS

Qwen3 235B A22B Instruct 2507

97.5 GB96.9 GB file

32K 100 GB · 128K 110 GB

Qwen3.8 27B

12.2 GB10.4 GB file

32K 13.6 GB · 128K 18.9 GB

IQ3_S

Qwen3 235B A22B Instruct 2507

not listed

Qwen3.8 27B

12.6 GB12.0 GB file

32K 13.9 GB · 128K 19.3 GB

IQ3_M

Qwen3 235B A22B Instruct 2507

106 GB107.3 GB file

32K 109 GB · 128K 119 GB

Qwen3.8 27B

13.3 GB14.9 GB file

32K 14.6 GB · 128K 20.0 GB

Q3_K_L

Qwen3 235B A22B Instruct 2507

121 GB111.2 GB file

32K 124 GB · 128K 133 GB

Qwen3.8 27B

15.0 GB14.1 GB file

32K 16.3 GB · 128K 21.7 GB

IQ4_XS

Qwen3 235B A22B Instruct 2507

127 GB125.5 GB file

32K 130 GB · 128K 139 GB

Qwen3.8 27B

15.7 GB14.3 GB file

32K 17.0 GB · 128K 22.4 GB

IQ4_NL

Qwen3 235B A22B Instruct 2507

133 GB133.1 GB file

32K 136 GB · 128K 145 GB

Qwen3.8 27B

16.4 GB17.4 GB file

32K 17.7 GB · 128K 23.1 GB

Q4_1

Qwen3 235B A22B Instruct 2507

133 GB147.2 GB file

32K 136 GB · 128K 145 GB

Qwen3.8 27B

16.4 GB17.5 GB file

32K 17.7 GB · 128K 23.1 GB

Q4_K_S

Qwen3 235B A22B Instruct 2507

133 GB133.7 GB file

32K 136 GB · 128K 145 GB

Qwen3.8 27B

16.4 GB15.4 GB file

32K 17.7 GB · 128K 23.1 GB

Q4_K_L

Qwen3 235B A22B Instruct 2507

144 GB143.1 GB file

32K 147 GB · 128K 157 GB

Qwen3.8 27B

17.8 GB18.8 GB file

32K 19.1 GB · 128K 24.5 GB

Q5_K_S

Qwen3 235B A22B Instruct 2507

162 GB161.9 GB file

32K 165 GB · 128K 175 GB

Qwen3.8 27B

19.9 GB18.7 GB file

32K 21.2 GB · 128K 26.6 GB

Q5_K_L

Qwen3 235B A22B Instruct 2507

not listed

Qwen3.8 27B

20.9 GB21.5 GB file

32K 22.2 GB · 128K 27.6 GB

BF16

Qwen3 235B A22B Instruct 2507

471 GB470.3 GB file

32K 474 GB · 128K 483 GB

Qwen3.8 27B

56.3 GB54.7 GB file

32K 57.7 GB · 128K 63.0 GB

How to run each

Qwen3 235B A22B Instruct 2507

Ollama · Q4_K_M

$ollama run hf.co/unsloth/Qwen3-235B-A22B-Instruct-2507-GGUF:Q4_K_M

All ways to run Qwen3 235B A22B Instruct 2507 →

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 →

Specifications

Specifications of Qwen3 235B A22B Instruct 2507 and Qwen3.8 27B
Spec	Qwen3 235B A22B Instruct 2507	Qwen3.8 27B
Publisher	Alibaba	Alibaba
Parameters	235.1B	27.8B
Active parameters	22B per token	—
Architecture	Qwen3MoeForCausalLM	Qwen3_5ForConditionalGeneration
Context	256K tokens	256K tokens
Modalities	Chat, Function-calling	Vision, Function-calling
License	Apache 2.0	Apache 2.0
Released	Jul 2025	Aug 2026

Frequently Asked Questions

Which is better, Qwen3 235B A22B Instruct 2507 or Qwen3.8 27B?

Qwen3.8 27B leads llmrun Score 121.5 to 98.5; only Qwen3.8 27B fits 24 GB (Q4_K_M); Qwen3 235B A22B Instruct 2507 needs 72.4 GB even at its smallest quant. Choose Qwen3 235B A22B Instruct 2507 for also does chat; choose Qwen3.8 27B for higher llmrun Score and stronger at reasoning.

Which needs less VRAM, Qwen3 235B A22B Instruct 2507 or Qwen3.8 27B?

At Q4_K_M, Qwen3 235B A22B Instruct 2507 requires 141.6 GB while Qwen3.8 27B needs 17.4 GB, making Qwen3.8 27B the easier model to run on your local setup.

Can I run Qwen3 235B A22B Instruct 2507 or Qwen3.8 27B on a 16 GB GPU?

Qwen3 235B A22B Instruct 2507 does not fit (smallest quant needs 72.4 GB); Qwen3.8 27B fits at Q3_K_L.

Which has a longer context window, Qwen3 235B A22B Instruct 2507 or Qwen3.8 27B?

Qwen3 235B A22B Instruct 2507 supports 262,144 tokens and Qwen3.8 27B supports 262,144 tokens.