DeepSeek R1 0528 vs Qwen3 Coder 480B A35B Instruct

DeepSeek R1 0528 and Qwen3 Coder 480B A35B Instruct match in llmrun Score, with values of 103.7 versus 103.9; neither fits 24 GB.

Model A · DeepSeek

DeepSeek R1 0528

103.7llmrun Score

684.5B · 40.1B active · 160K context · MIT · May 2025

ChatReasoningFunctions

Model B · Alibaba

Qwen3 Coder 480B A35B Instruct

103.9llmrun Score

480.2B · 35.5B active · 256K context · Apache 2.0 · Jul 2025

ChatCodeFunctions

Which should you choose?

Choose DeepSeek R1 0528 if you want…

Stronger at coding
107.3 vs 100.6
Also does reasoning
Qwen3 Coder 480B A35B Instruct doesn't list it

Choose Qwen3 Coder 480B A35B Instruct if you want…

Lighter on memory
288.7 vs 414.6 GB at Q4_K_M
Longer context window
256K vs 160K tokens
Also does code
DeepSeek R1 0528 doesn't list it

On your hardware

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

MemoryDeepSeek R1 0528Qwen3 Coder 480B A35B Instruct

8 GB

RTX 4060 · 8 GB tier

DeepSeek R1 0528

Doesn't fit

smallest quant needs 189.7 GB

Qwen3 Coder 480B A35B Instruct

Doesn't fit

smallest quant needs 146.4 GB

12 GB

RTX 3060 12GB · 12 GB tier

DeepSeek R1 0528

Doesn't fit

smallest quant needs 189.7 GB

Qwen3 Coder 480B A35B Instruct

Doesn't fit

smallest quant needs 146.4 GB

16 GB

RTX 5060 Ti 16GB · 16 GB tier

DeepSeek R1 0528

Doesn't fit

smallest quant needs 189.7 GB

Qwen3 Coder 480B A35B Instruct

Doesn't fit

smallest quant needs 146.4 GB

24 GBreference

RTX 4090 · 24 GB tier

DeepSeek R1 0528

Doesn't fit

smallest quant needs 189.7 GB

Qwen3 Coder 480B A35B Instruct

Doesn't fit

smallest quant needs 146.4 GB

32 GB

RTX 5090 · 32 GB tier

DeepSeek R1 0528

Doesn't fit

smallest quant needs 189.7 GB

Qwen3 Coder 480B A35B Instruct

Doesn't fit

smallest quant needs 146.4 GB

48 GB

RTX A6000 · 48 GB tier

DeepSeek R1 0528

Doesn't fit

smallest quant needs 189.7 GB

Qwen3 Coder 480B A35B Instruct

Doesn't fit

smallest quant needs 146.4 GB

Mac 32 GB

Mac Mini M4 · Macs

DeepSeek R1 0528

Doesn't fit

smallest quant needs 189.7 GB

Qwen3 Coder 480B A35B Instruct

Doesn't fit

smallest quant needs 146.4 GB

Mac 64 GB

Mac Studio M4 Max · Macs

DeepSeek R1 0528

Doesn't fit

smallest quant needs 189.7 GB

Qwen3 Coder 480B A35B Instruct

Doesn't fit

smallest quant needs 146.4 GB

Mac 128 GB

Mac Studio M4 Max · Macs

DeepSeek R1 0528

Doesn't fit

smallest quant needs 189.7 GB

Qwen3 Coder 480B A35B Instruct

Doesn't fit

smallest quant needs 146.4 GB

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

DeepSeek R1 0528 Qwen3 Coder 480B A35B Instruct

Overall

DeepSeek R1 0528

103.7101–106

Qwen3 Coder 480B A35B Instruct

103.9100–108

Coding

DeepSeek R1 0528

107.3102–113

Qwen3 Coder 480B A35B Instruct

100.691–110

Agents

DeepSeek R1 0528

no data

Qwen3 Coder 480B A35B Instruct

104.9100–110

Reasoning

DeepSeek R1 0528

100.094–106

Qwen3 Coder 480B A35B Instruct

no data

DeepSeek R1 0528 · #44 of 100 open models · #41 of 96 that fit ≤512 GB

Qwen3 Coder 480B A35B Instruct · #42 of 100 open models · #39 of 96 that fit ≤512 GB

Benchmarks

2 shared boards: DeepSeek R1 0528 leads 2, Qwen3 Coder 480B A35B Instruct leads 0 · Benchmarks imported Oct 2026

BoardDeepSeek R1 0528Qwen3 Coder 480B A35B Instruct

WeirdML

DeepSeek R1 0528

41.6%#18/46 open

IndependentEpoch AI · May 2025

Qwen3 Coder 480B A35B Instruct

41.2%#19/46 open

IndependentEpoch AI · Jul 2025

ALE-Bench

DeepSeek R1 0528

804.1#18/43 open

Lab-reportedEpoch AI · May 2025

Qwen3 Coder 480B A35B Instruct

461.4#33/43 open

Lab-reportedEpoch AI · Jul 2025

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

LMArena Text

DeepSeek R1 0528

1421.4#33/187 open

ArenaLMArena · Oct 2026

Qwen3 Coder 480B A35B Instruct

1387.6#58/187 open

ArenaLMArena · Oct 2026

Only DeepSeek R1 0528 (10)

Only Qwen3 Coder 480B A35B Instruct (5)

VRAM by quantization

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

QuantDeepSeek R1 0528Qwen3 Coder 480B A35B Instruct

Q2_K

DeepSeek R1 0528

295 GB237.8 GB file

32K 294 GB · 128K 300 GB

Qwen3 Coder 480B A35B Instruct

205 GB174.7 GB file

32K 209 GB · 128K 221 GB

Q3_K_S

DeepSeek R1 0528

303 GB293.4 GB file

32K 302 GB · 128K 309 GB

Qwen3 Coder 480B A35B Instruct

211 GB206.9 GB file

32K 215 GB · 128K 227 GB

Q3_K_M

DeepSeek R1 0528

338 GB307.9 GB file

32K 336 GB · 128K 343 GB

Qwen3 Coder 480B A35B Instruct

235 GB229.2 GB file

32K 239 GB · 128K 251 GB

Q4_0

DeepSeek R1 0528

346 GB386.4 GB file

32K 345 GB · 128K 352 GB

Qwen3 Coder 480B A35B Instruct

241 GB271.6 GB file

32K 245 GB · 128K 257 GB

Q4_K_M

DeepSeek R1 0528

415 GB409.2 GB file

32K 413 GB · 128K 420 GB

Qwen3 Coder 480B A35B Instruct

289 GB290.1 GB file

32K 293 GB · 128K 305 GB

Q5_K_M

DeepSeek R1 0528

492 GB478.3 GB file

32K 490 GB · 128K 497 GB

Qwen3 Coder 480B A35B Instruct

343 GB340.5 GB file

32K 347 GB · 128K 359 GB

Q6_K

DeepSeek R1 0528

569 GB552.4 GB file

32K 567 GB · 128K 574 GB

Qwen3 Coder 480B A35B Instruct

397 GB394.1 GB file

32K 401 GB · 128K 413 GB

Q8_0

DeepSeek R1 0528

688 GB713.3 GB file

32K 687 GB · 128K 694 GB

Qwen3 Coder 480B A35B Instruct

481 GB510.4 GB file

32K 485 GB · 128K 497 GB

14 more quants

IQ2_XXS

DeepSeek R1 0528

192 GB164.1 GB file

32K 191 GB · 128K 198 GB

Qwen3 Coder 480B A35B Instruct

not listed

IQ2_XS

DeepSeek R1 0528

209 GB188.4 GB file

32K 208 GB · 128K 215 GB

Qwen3 Coder 480B A35B Instruct

145 GB132.9 GB file

32K 149 GB · 128K 161 GB

IQ2_S

DeepSeek R1 0528

218 GB189.6 GB file

32K 217 GB · 128K 223 GB

Qwen3 Coder 480B A35B Instruct

151 GB133.6 GB file

32K 155 GB · 128K 167 GB

IQ2_M

DeepSeek R1 0528

235 GB215.0 GB file

32K 234 GB · 128K 241 GB

Qwen3 Coder 480B A35B Instruct

not listed

IQ3_XXS

DeepSeek R1 0528

269 GB267.6 GB file

32K 268 GB · 128K 275 GB

Qwen3 Coder 480B A35B Instruct

187 GB201.5 GB file

32K 191 GB · 128K 203 GB

IQ3_XS

DeepSeek R1 0528

286 GB277.1 GB file

32K 285 GB · 128K 292 GB

Qwen3 Coder 480B A35B Instruct

199 GB197.1 GB file

32K 203 GB · 128K 215 GB

IQ3_M

DeepSeek R1 0528

312 GB307.9 GB file

32K 311 GB · 128K 318 GB

Qwen3 Coder 480B A35B Instruct

217 GB218.4 GB file

32K 221 GB · 128K 233 GB

Q3_K_L

DeepSeek R1 0528

355 GB319.7 GB file

32K 353 GB · 128K 360 GB

Qwen3 Coder 480B A35B Instruct

247 GB226.7 GB file

32K 251 GB · 128K 263 GB

IQ4_XS

DeepSeek R1 0528

372 GB360.0 GB file

32K 371 GB · 128K 377 GB

Qwen3 Coder 480B A35B Instruct

259 GB260.5 GB file

32K 263 GB · 128K 275 GB

IQ4_NL

DeepSeek R1 0528

389 GB380.5 GB file

32K 388 GB · 128K 395 GB

Qwen3 Coder 480B A35B Instruct

271 GB270.6 GB file

32K 275 GB · 128K 287 GB

Q4_1

DeepSeek R1 0528

389 GB421.0 GB file

32K 388 GB · 128K 395 GB

Qwen3 Coder 480B A35B Instruct

271 GB300.5 GB file

32K 275 GB · 128K 287 GB

Q4_K_S

DeepSeek R1 0528

389 GB394.2 GB file

32K 388 GB · 128K 395 GB

Qwen3 Coder 480B A35B Instruct

271 GB272.8 GB file

32K 275 GB · 128K 287 GB

Q5_K_S

DeepSeek R1 0528

475 GB463.0 GB file

32K 473 GB · 128K 480 GB

Qwen3 Coder 480B A35B Instruct

331 GB330.4 GB file

32K 335 GB · 128K 347 GB

BF16

DeepSeek R1 0528

1373 GB1342.3 GB file

32K 1372 GB · 128K 1379 GB

Qwen3 Coder 480B A35B Instruct

961 GB960.4 GB file

32K 965 GB · 128K 977 GB

How to run each

DeepSeek R1 0528

Ollama · Q4_K_M

$ollama run hf.co/bartowski/deepseek-ai_DeepSeek-R1-0528-GGUF:Q4_K_M

All ways to run DeepSeek R1 0528 →

Qwen3 Coder 480B A35B Instruct

Ollama · Q4_K_M

$ollama run hf.co/unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF:Q4_K_M

All ways to run Qwen3 Coder 480B A35B Instruct →

Specifications

Specifications of DeepSeek R1 0528 and Qwen3 Coder 480B A35B Instruct
Spec	DeepSeek R1 0528	Qwen3 Coder 480B A35B Instruct
Publisher	DeepSeek	Alibaba
Parameters	684.5B	480.2B
Active parameters	40B per token	35B per token
Architecture	DeepseekV3ForCausalLM	Qwen3MoeForCausalLM
Context	160K tokens	256K tokens
Modalities	Chat, Reasoning, Function-calling	Chat, Code, Function-calling
License	MIT	Apache 2.0
Released	May 2025	Jul 2025

Frequently Asked Questions

Which is better, DeepSeek R1 0528 or Qwen3 Coder 480B A35B Instruct?

DeepSeek R1 0528 and Qwen3 Coder 480B A35B Instruct match in llmrun Score, with values of 103.7 versus 103.9; neither fits 24 GB. Choose DeepSeek R1 0528 for stronger at coding and also does reasoning; choose Qwen3 Coder 480B A35B Instruct for lighter on memory and longer context window.

Which needs less VRAM, DeepSeek R1 0528 or Qwen3 Coder 480B A35B Instruct?

At Q4_K_M, DeepSeek R1 0528 requires 414.6 GB while Qwen3 Coder 480B A35B Instruct needs 288.7 GB, making Qwen3 Coder 480B A35B Instruct the easier model to run on your local setup.

Can I run DeepSeek R1 0528 or Qwen3 Coder 480B A35B Instruct on a 16 GB GPU?

DeepSeek R1 0528 does not fit (smallest quant needs 189.7 GB); Qwen3 Coder 480B A35B Instruct does not fit (smallest quant needs 146.4 GB).

Which has a longer context window, DeepSeek R1 0528 or Qwen3 Coder 480B A35B Instruct?

DeepSeek R1 0528 supports 163,840 tokens and Qwen3 Coder 480B A35B Instruct supports 262,144 tokens.