MiniMax M2.5 vs Qwen3.5 397B A17B
MiniMax M2.5 and Qwen3.5 397B A17B match in llmrun Score, with values of 111.8 versus 112.6; neither fits 24 GB.
Model A · MiniMax
MiniMax M2.5228.7B · 11B active · 192K context · Other · Feb 2026
Model B · Alibaba
Qwen3.5 397B A17B403.4B · 24.4B active · 256K context · Apache 2.0 · Feb 2026
Choose MiniMax M2.5 if you want…
Fits a Mac 128 GB
Q3_K_L · 119.6 GB; Qwen3.5 397B A17B needs 111.5 GB at its smallest quant
Lighter on memory
137.8 vs 242.5 GB at Q4_K_M
Also does chat
Qwen3.5 397B A17B doesn't list it
Choose Qwen3.5 397B A17B if you want…
Ahead on the benchmarks
3 of 3 shared boards
Longer context window
256K vs 192K tokens
Also does vision
MiniMax M2.5 doesn't list it
More permissive license
Apache 2.0 vs Other
On your hardware
Each model at its default quant (Q4_K_M), a smaller one when it must. Estimates, ±20%.
8 GB
RTX 4060 · 8 GB tier
MiniMax M2.5
Doesn't fit
smallest quant needs 65.3 GB
Qwen3.5 397B A17B
Doesn't fit
smallest quant needs 111.5 GB
12 GB
RTX 3060 12GB · 12 GB tier
MiniMax M2.5
Doesn't fit
smallest quant needs 65.3 GB
Qwen3.5 397B A17B
Doesn't fit
smallest quant needs 111.5 GB
16 GB
RTX 5060 Ti 16GB · 16 GB tier
MiniMax M2.5
Doesn't fit
smallest quant needs 65.3 GB
Qwen3.5 397B A17B
Doesn't fit
smallest quant needs 111.5 GB
24 GBreference
RTX 4090 · 24 GB tier
MiniMax M2.5
Doesn't fit
smallest quant needs 65.3 GB
Qwen3.5 397B A17B
Doesn't fit
smallest quant needs 111.5 GB
32 GB
RTX 5090 · 32 GB tier
MiniMax M2.5
Doesn't fit
smallest quant needs 65.3 GB
Qwen3.5 397B A17B
Doesn't fit
smallest quant needs 111.5 GB
48 GB
RTX A6000 · 48 GB tier
MiniMax M2.5
Doesn't fit
smallest quant needs 65.3 GB
Qwen3.5 397B A17B
Doesn't fit
smallest quant needs 111.5 GB
Mac 32 GB
Mac Mini M4 · Macs
MiniMax M2.5
Doesn't fit
smallest quant needs 65.3 GB
Qwen3.5 397B A17B
Doesn't fit
smallest quant needs 111.5 GB
Mac 64 GB
Mac Studio M4 Max · Macs
MiniMax M2.5
Doesn't fit
smallest quant needs 65.3 GB
Qwen3.5 397B A17B
Doesn't fit
smallest quant needs 111.5 GB
Mac 128 GB
Mac Studio M4 Max · Macs
MiniMax M2.5
Q3_K_L · ~37 tok/s
119.6 GB · up to 32Kbetter quant
Qwen3.5 397B A17B
IQ2_XS · ~32 tok/s
121.6 GB · up to 128K
Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).
Overall
MiniMax M2.5
Qwen3.5 397B A17B
Coding
MiniMax M2.5
Qwen3.5 397B A17B
no data
Agents
MiniMax M2.5
Qwen3.5 397B A17B
no data
Math
MiniMax M2.5
no data
Qwen3.5 397B A17B
Reasoning
MiniMax M2.5
Qwen3.5 397B A17B
MiniMax M2.5 · #19 of 92 open models · #6 of 71 that fit ≤192 GB
Qwen3.5 397B A17B · #18 of 92 open models · #8 of 77 that fit ≤256 GB
Benchmarks
3 shared boards: MiniMax M2.5 leads 0, Qwen3.5 397B A17B leads 3 · Benchmarks imported Oct 2026
MiniMax M2.5
1390.9#56/186 open
ArenaLMArena · Oct 2026
Qwen3.5 397B A17B
1441.7#17/186 open
ArenaLMArena · Oct 2026
MiniMax M2.5
1386.7#28/50 open
ArenaLMArena · Oct 2026
Qwen3.5 397B A17B
1399.9#26/50 open
ArenaLMArena · Oct 2026
MiniMax M2.5
80.1%#25/144 open
IndependentTIGER-Lab
Qwen3.5 397B A17B
87.8%#2/144 open
IndependentTIGER-Lab
18 more boards (other shared and single-model boards)
Only MiniMax M2.5 (10)
- SWE-bench Verified75.8%
- Terminal-Bench42.7%
- ARC-AGI-24.9%
- SWE-bench Multilingual68.3%
- ARC-AGI63.7%
- CL-bench11.4%
- CL-bench Life6.3%
- Vending-Bench 2-23.2
- ALE-Bench618.2
- ProofBench4.0%
Only Qwen3.5 397B A17B (8)
- GPQA Diamond86.4%
- AIME 2024/202588.9%
- FrontierMath31.2%
- Chess Puzzles13.0%
- DTBench87.5%
- LMCA37.9%
- APEX-Agents24.9%
- Mystery Game Puzzles18.0%
VRAM by quantization
Memory to load the model and 2K context; below it, the total at 32K and 128K context. Lighter side in bold.
Q2_K
MiniMax M2.5
97.8 GB83.3 GB file
32K 102 GB · 128K 114 GB
Qwen3.5 397B A17B
172 GB146.3 GB file
32K 172 GB · 128K 174 GB
Q3_K_S
MiniMax M2.5
101 GB98.7 GB file
32K 105 GB · 128K 117 GB
Qwen3.5 397B A17B
177 GB164.3 GB file
32K 177 GB · 128K 179 GB
Q3_K_M
MiniMax M2.5
112 GB109.3 GB file
32K 116 GB · 128K 128 GB
Qwen3.5 397B A17B
197 GB177.4 GB file
32K 197 GB · 128K 199 GB
Q4_0
MiniMax M2.5
115 GB129.5 GB file
32K 119 GB · 128K 131 GB
Qwen3.5 397B A17B
202 GB~201.7 GB file
32K 203 GB · 128K 204 GB
Q4_K_M
MiniMax M2.5
138 GB138.3 GB file
32K 142 GB · 128K 154 GB
Qwen3.5 397B A17B
242 GB244.1 GB file
32K 243 GB · 128K 244 GB
Q5_K_M
MiniMax M2.5
164 GB162.3 GB file
32K 167 GB · 128K 180 GB
Qwen3.5 397B A17B
288 GB293.7 GB file
32K 288 GB · 128K 290 GB
Q6_K
MiniMax M2.5
189 GB187.8 GB file
32K 193 GB · 128K 206 GB
Qwen3.5 397B A17B
333 GB326.6 GB file
32K 334 GB · 128K 335 GB
Q8_0
MiniMax M2.5
229 GB243.1 GB file
32K 233 GB · 128K 246 GB
Qwen3.5 397B A17B
404 GB421.5 GB file
32K 404 GB · 128K 406 GB
18 more quants
IQ2_XXS
MiniMax M2.5
63.5 GB74.1 GB file
32K 67.4 GB · 128K 79.8 GB
Qwen3.5 397B A17B
111 GB114.9 GB file
32K 112 GB · 128K 113 GB
IQ2_XS
MiniMax M2.5
69.2 GB67.0 GB file
32K 73.1 GB · 128K 85.6 GB
Qwen3.5 397B A17B
121 GB125.4 GB file
32K 122 GB · 128K 123 GB
IQ2_S
MiniMax M2.5
72.0 GB68.0 GB file
32K 75.9 GB · 128K 88.4 GB
Qwen3.5 397B A17B
126 GB127.3 GB file
32K 127 GB · 128K 128 GB
IQ2_M
MiniMax M2.5
77.8 GB78.2 GB file
32K 81.7 GB · 128K 94.1 GB
Qwen3.5 397B A17B
137 GB123.1 GB file
32K 137 GB · 128K 138 GB
IQ3_XXS
MiniMax M2.5
89.2 GB93.3 GB file
32K 93.1 GB · 128K 106 GB
Qwen3.5 397B A17B
157 GB140.3 GB file
32K 157 GB · 128K 159 GB
Q2_K_S
MiniMax M2.5
92.0 GB77.6 GB file
32K 95.9 GB · 128K 108 GB
Qwen3.5 397B A17B
not listed
IQ3_XS
MiniMax M2.5
94.9 GB93.3 GB file
32K 98.8 GB · 128K 111 GB
Qwen3.5 397B A17B
167 GB188.2 GB file
32K 167 GB · 128K 169 GB
IQ3_S
MiniMax M2.5
97.8 GB98.7 GB file
32K 102 GB · 128K 114 GB
Qwen3.5 397B A17B
172 GB146.4 GB file
32K 172 GB · 128K 174 GB
IQ3_M
MiniMax M2.5
103 GB100.0 GB file
32K 107 GB · 128K 120 GB
Qwen3.5 397B A17B
182 GB196.0 GB file
32K 182 GB · 128K 184 GB
Q3_K_L
MiniMax M2.5
118 GB118.4 GB file
32K 122 GB · 128K 134 GB
Qwen3.5 397B A17B
207 GB195.9 GB file
32K 208 GB · 128K 209 GB
IQ4_XS
MiniMax M2.5
123 GB121.9 GB file
32K 127 GB · 128K 140 GB
Qwen3.5 397B A17B
217 GB189.7 GB file
32K 218 GB · 128K 219 GB
IQ4_NL
MiniMax M2.5
129 GB129.0 GB file
32K 133 GB · 128K 146 GB
Qwen3.5 397B A17B
227 GB193.8 GB file
32K 228 GB · 128K 229 GB
Q4_1
MiniMax M2.5
129 GB143.2 GB file
32K 133 GB · 128K 146 GB
Qwen3.5 397B A17B
227 GB256.1 GB file
32K 228 GB · 128K 229 GB
Q4_K_S
MiniMax M2.5
129 GB130.0 GB file
32K 133 GB · 128K 146 GB
Qwen3.5 397B A17B
227 GB228.0 GB file
32K 228 GB · 128K 229 GB
Q4_K_L
MiniMax M2.5
not listed
Qwen3.5 397B A17B
248 GB249.0 GB file
32K 248 GB · 128K 249 GB
Q5_K_S
MiniMax M2.5
158 GB157.5 GB file
32K 162 GB · 128K 174 GB
Qwen3.5 397B A17B
278 GB276.6 GB file
32K 278 GB · 128K 280 GB
Q5_K_L
MiniMax M2.5
not listed
Qwen3.5 397B A17B
293 GB290.6 GB file
32K 293 GB · 128K 295 GB
BF16
MiniMax M2.5
458 GB457.5 GB file
32K 462 GB · 128K 474 GB
Qwen3.5 397B A17B
807 GB793.0 GB file
32K 808 GB · 128K 809 GB
MiniMax M2.5
Ollama · Q4_K_M
ollama run hf.co/unsloth/MiniMax-M2.5-GGUF:Q4_K_MQwen3.5 397B A17B
Ollama · Q4_K_M
ollama run hf.co/unsloth/Qwen3.5-397B-A17B-GGUF:Q4_K_M| Spec | MiniMax M2.5 | Qwen3.5 397B A17B |
|---|---|---|
| Publisher | MiniMax | Alibaba |
| Parameters | 228.7B | 403.4B |
| Active parameters | 11B per token | 24B per token |
| Architecture | MiniMaxM2ForCausalLM | Qwen3_5MoeForConditionalGeneration |
| Context | 192K tokens | 256K tokens |
| Modalities | Chat, Function-calling | Vision, Function-calling |
| License | Other | Apache 2.0 |
| Released | Feb 2026 | Feb 2026 |
Frequently Asked Questions
Which is better, MiniMax M2.5 or Qwen3.5 397B A17B?
MiniMax M2.5 and Qwen3.5 397B A17B match in llmrun Score, with values of 111.8 versus 112.6; neither fits 24 GB. Choose MiniMax M2.5 for fits a Mac 128 GB and lighter on memory; choose Qwen3.5 397B A17B for ahead on the benchmarks and longer context window.
Which needs less VRAM, MiniMax M2.5 or Qwen3.5 397B A17B?
At Q4_K_M, MiniMax M2.5 requires 137.8 GB while Qwen3.5 397B A17B needs 242.5 GB, making MiniMax M2.5 the easier model to run on your local setup.
Can I run MiniMax M2.5 or Qwen3.5 397B A17B on a 16 GB GPU?
MiniMax M2.5 does not fit (smallest quant needs 65.3 GB); Qwen3.5 397B A17B does not fit (smallest quant needs 111.5 GB).
Which has a longer context window, MiniMax M2.5 or Qwen3.5 397B A17B?
MiniMax M2.5 supports 196,608 tokens and Qwen3.5 397B A17B supports 262,144 tokens.