DeepSeek V3.1 vs GLM 4.5
DeepSeek V3.1 and GLM 4.5 match in llmrun Score, with values of 99.8 versus 100.9; neither fits 24 GB.
Model A · DeepSeek
DeepSeek V3.1684.5B · 40.1B active · 160K context · MIT · Aug 2025
Model B · Z.ai
GLM 4.5358.3B · 35.6B active · 128K context · MIT · Jul 2025
Choose GLM 4.5 if you want…
Lighter on memory
215.6 vs 414.6 GB at Q4_K_M
DeepSeek V3.1: no clear edge over GLM 4.5 in the data we have.
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
DeepSeek V3.1
Doesn't fit
smallest quant needs 189.7 GB
GLM 4.5
Doesn't fit
smallest quant needs 101.4 GB
12 GB
RTX 3060 12GB · 12 GB tier
DeepSeek V3.1
Doesn't fit
smallest quant needs 189.7 GB
GLM 4.5
Doesn't fit
smallest quant needs 101.4 GB
16 GB
RTX 5060 Ti 16GB · 16 GB tier
DeepSeek V3.1
Doesn't fit
smallest quant needs 189.7 GB
GLM 4.5
Doesn't fit
smallest quant needs 101.4 GB
24 GBreference
RTX 4090 · 24 GB tier
DeepSeek V3.1
Doesn't fit
smallest quant needs 189.7 GB
GLM 4.5
Doesn't fit
smallest quant needs 101.4 GB
32 GB
RTX 5090 · 32 GB tier
DeepSeek V3.1
Doesn't fit
smallest quant needs 189.7 GB
GLM 4.5
Doesn't fit
smallest quant needs 101.4 GB
48 GB
RTX A6000 · 48 GB tier
DeepSeek V3.1
Doesn't fit
smallest quant needs 189.7 GB
GLM 4.5
Doesn't fit
smallest quant needs 101.4 GB
Mac 32 GB
Mac Mini M4 · Macs
DeepSeek V3.1
Doesn't fit
smallest quant needs 189.7 GB
GLM 4.5
Doesn't fit
smallest quant needs 101.4 GB
Mac 64 GB
Mac Studio M4 Max · Macs
DeepSeek V3.1
Doesn't fit
smallest quant needs 189.7 GB
GLM 4.5
Doesn't fit
smallest quant needs 101.4 GB
Mac 128 GB
Mac Studio M4 Max · Macs
DeepSeek V3.1
Doesn't fit
smallest quant needs 189.7 GB
GLM 4.5
IQ2_M · ~20 tok/s
123.8 GB · up to 16Kfits
Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).
Overall
DeepSeek V3.1
GLM 4.5
Coding
DeepSeek V3.1
no data
GLM 4.5
Agents
DeepSeek V3.1
no data
GLM 4.5
Reasoning
DeepSeek V3.1
GLM 4.5
no data
DeepSeek V3.1 · #47 of 92 open models · #44 of 88 that fit ≤512 GB
GLM 4.5 · #44 of 92 open models · #31 of 77 that fit ≤256 GB
DeepSeek V3.1
1417.2#36/186 open
ArenaLMArena · Oct 2026
GLM 4.5
1411.0#42/186 open
ArenaLMArena · Oct 2026
DeepSeek V3.1
84.8%#9/144 open
IndependentTIGER-Lab
GLM 4.5
84.6%#10/144 open
IndependentTIGER-Lab
DeepSeek V3.1
38.4%#27/46 open
IndependentEpoch AI · Aug 2025
GLM 4.5
40.6%#22/46 open
IndependentEpoch AI · Aug 2025
9 more boards (other shared and single-model boards)
DeepSeek V3.1
52.8%#17/25 open
IndependentEpoch AI · Aug 2025
GLM 4.5
58.3%#14/25 open
IndependentEpoch AI · Jul 2025
Only DeepSeek V3.1 (4)
- SimpleBench40.0%
- DTBench82.7%
- LMCA24.3%
- ForecastBench58.0
Only GLM 4.5 (4)
- SWE-bench Verified64.2%
- Humanity's Last Exam8.3%
- SWE-bench Bash Only54.2%
- ALE-Bench344.8
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
DeepSeek V3.1
295 GB245.7 GB file
32K 294 GB · 128K 300 GB
GLM 4.5
153 GB131.2 GB file
32K 158 GB · 128K 173 GB
Q3_K_S
DeepSeek V3.1
303 GB290.4 GB file
32K 302 GB · 128K 309 GB
GLM 4.5
157 GB155.0 GB file
32K 162 GB · 128K 178 GB
Q3_K_M
DeepSeek V3.1
338 GB320.5 GB file
32K 336 GB · 128K 343 GB
GLM 4.5
175 GB171.3 GB file
32K 180 GB · 128K 196 GB
Q4_0
DeepSeek V3.1
346 GB380.1 GB file
32K 345 GB · 128K 352 GB
GLM 4.5
180 GB203.0 GB file
32K 185 GB · 128K 200 GB
Q4_K_M
DeepSeek V3.1
415 GB405.4 GB file
32K 413 GB · 128K 420 GB
GLM 4.5
216 GB216.5 GB file
32K 220 GB · 128K 236 GB
Q5_K_M
DeepSeek V3.1
492 GB476.2 GB file
32K 490 GB · 128K 497 GB
GLM 4.5
256 GB254.2 GB file
32K 261 GB · 128K 276 GB
Q6_K
DeepSeek V3.1
569 GB551.3 GB file
32K 567 GB · 128K 574 GB
GLM 4.5
296 GB294.4 GB file
32K 301 GB · 128K 317 GB
Q8_0
DeepSeek V3.1
688 GB713.3 GB file
32K 687 GB · 128K 694 GB
GLM 4.5
359 GB381.0 GB file
32K 364 GB · 128K 379 GB
14 more quants
IQ2_XXS
DeepSeek V3.1
192 GB225.9 GB file
32K 191 GB · 128K 198 GB
GLM 4.5
99.2 GB115.9 GB file
32K 104 GB · 128K 119 GB
IQ2_XS
DeepSeek V3.1
209 GB188.4 GB file
32K 208 GB · 128K 215 GB
GLM 4.5
108 GB101.8 GB file
32K 113 GB · 128K 128 GB
IQ2_S
DeepSeek V3.1
218 GB189.6 GB file
32K 217 GB · 128K 223 GB
GLM 4.5
113 GB102.7 GB file
32K 117 GB · 128K 133 GB
IQ2_M
DeepSeek V3.1
235 GB235.5 GB file
32K 234 GB · 128K 241 GB
GLM 4.5
122 GB122.4 GB file
32K 126 GB · 128K 142 GB
IQ3_XXS
DeepSeek V3.1
269 GB279.7 GB file
32K 268 GB · 128K 275 GB
GLM 4.5
139 GB145.2 GB file
32K 144 GB · 128K 160 GB
IQ3_XS
DeepSeek V3.1
286 GB277.1 GB file
32K 285 GB · 128K 292 GB
GLM 4.5
148 GB148.9 GB file
32K 153 GB · 128K 169 GB
IQ3_M
DeepSeek V3.1
312 GB307.9 GB file
32K 311 GB · 128K 318 GB
GLM 4.5
162 GB164.9 GB file
32K 167 GB · 128K 182 GB
Q3_K_L
DeepSeek V3.1
355 GB319.7 GB file
32K 353 GB · 128K 360 GB
GLM 4.5
184 GB170.6 GB file
32K 189 GB · 128K 205 GB
IQ4_XS
DeepSeek V3.1
372 GB358.3 GB file
32K 371 GB · 128K 377 GB
GLM 4.5
193 GB191.6 GB file
32K 198 GB · 128K 213 GB
IQ4_NL
DeepSeek V3.1
389 GB379.1 GB file
32K 388 GB · 128K 395 GB
GLM 4.5
202 GB202.5 GB file
32K 207 GB · 128K 222 GB
Q4_1
DeepSeek V3.1
389 GB420.8 GB file
32K 388 GB · 128K 395 GB
GLM 4.5
202 GB224.6 GB file
32K 207 GB · 128K 222 GB
Q4_K_S
DeepSeek V3.1
389 GB381.1 GB file
32K 388 GB · 128K 395 GB
GLM 4.5
202 GB203.7 GB file
32K 207 GB · 128K 222 GB
Q5_K_S
DeepSeek V3.1
475 GB462.6 GB file
32K 473 GB · 128K 480 GB
GLM 4.5
247 GB247.0 GB file
32K 252 GB · 128K 267 GB
BF16
DeepSeek V3.1
1373 GB1342.3 GB file
32K 1372 GB · 128K 1379 GB
GLM 4.5
717 GB716.8 GB file
32K 722 GB · 128K 738 GB
DeepSeek V3.1
Ollama · Q4_K_M
ollama run hf.co/unsloth/DeepSeek-V3.1-GGUF:Q4_K_MGLM 4.5
Ollama · Q4_K_M
ollama run hf.co/unsloth/GLM-4.5-GGUF:Q4_K_M| Spec | DeepSeek V3.1 | GLM 4.5 |
|---|---|---|
| Publisher | DeepSeek | Z.ai |
| Parameters | 684.5B | 358.3B |
| Active parameters | 40B per token | 36B per token |
| Architecture | DeepseekV3ForCausalLM | Glm4MoeForCausalLM |
| Context | 160K tokens | 128K tokens |
| Modalities | Chat, Function-calling | Chat, Function-calling |
| License | MIT | MIT |
| Released | Aug 2025 | Jul 2025 |
Frequently Asked Questions
Which is better, DeepSeek V3.1 or GLM 4.5?
DeepSeek V3.1 and GLM 4.5 match in llmrun Score, with values of 99.8 versus 100.9; neither fits 24 GB. Choose GLM 4.5 for lighter on memory.
Which needs less VRAM, DeepSeek V3.1 or GLM 4.5?
At Q4_K_M, DeepSeek V3.1 requires 414.6 GB while GLM 4.5 needs 215.6 GB, making GLM 4.5 the easier model to run on your local setup.
Can I run DeepSeek V3.1 or GLM 4.5 on a 16 GB GPU?
DeepSeek V3.1 does not fit (smallest quant needs 189.7 GB); GLM 4.5 does not fit (smallest quant needs 101.4 GB).
Which has a longer context window, DeepSeek V3.1 or GLM 4.5?
DeepSeek V3.1 supports 163,840 tokens and GLM 4.5 supports 131,072 tokens.