Noometry

Model comparison

DeepSeek V4 Pro vs Qwen3.5 27B

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 41.9 on the Noometry Index.

Last verified . 26 shared benchmarks.

DeepSeek V4 Pro DeepSeek

54.3

Rank #31 Confirmed

Qwen3.5 27B Alibaba (Qwen)

41.9

Rank #127 Confirmed

Summary

  • They share 26 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Qwen3.5 27B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 27.5.
  • The biggest single-benchmark swing is NYT Connections (extended): 91.3% for DeepSeek V4 Pro and 47.9% for Qwen3.5 27B.
  • Qwen3.5 27B is cheaper at $0.30 / $2.40 per million input/output tokens, against $0.66 / $1.98 for DeepSeek V4 Pro.
  • DeepSeek V4 Pro accepts more context: 1M tokens versus 262K.

Side by side

DeepSeek V4 ProQwen3.5 27B
ProviderDeepSeekAlibaba (Qwen)
Noometry Index54.341.9
Released2026-04-242026-02-23
WeightsOpenOpen
Context window1M262K
Max output393K66K
Input $ / M tokens$0.66$0.30
Output $ / M tokens$1.98$2.40
Results tracked4828

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding DeepSeek V4 Pro leads

DeepSeek V4 Pro: 52.4 (#34), Qwen3.5 27B: 38.9 (#168)

BenchmarkDeepSeek V4 ProQwen3.5 27B
LMArena WebDev15821358
WeirdML66.2%39.5%
LMArena Coding14701427
ALE-Bench1,403349.45
SWE-bench Verified77.6%—
FrontierCode28.6%—
SciCode51%—

Agentic & Tool Use Not comparable

DeepSeek V4 Pro: 32.8 (#58), Qwen3.5 27B: —

BenchmarkDeepSeek V4 ProQwen3.5 27B
Vending-Bench 23,285201.98
APEX-Agents47.3%—

Reasoning DeepSeek V4 Pro leads

DeepSeek V4 Pro: 56.5 (#24), Qwen3.5 27B: 27.5 (#117)

BenchmarkDeepSeek V4 ProQwen3.5 27B
NYT Connections (extended)91.3%47.9%
LMArena Hard Prompts14611414
DTBench93.9%82.4%
LMCA45.5%34%
ARC-AGI-261.3%—
Kagi LLM Benchmark53.5%—
ARC-AGI-190.5%—
CritPt18%—
Chess Puzzles47%—
Thematic Generalization—45.5%
Mystery Game Puzzles43%—
Surface Evolver Bench40%—
Epoch Capabilities Index155.31—
ForecastBench56.1—

Math DeepSeek V4 Pro leads

DeepSeek V4 Pro: 64.8 (#30), Qwen3.5 27B: 38.8 (#127)

BenchmarkDeepSeek V4 ProQwen3.5 27B
MathArena Final-Answer Competitions76.6%56.7%
LMArena Math14551429
FrontierMath (Tiers 1-3)64.6%—
FrontierMath Tier 426.8%—
OTIS Mock AIME 2024-202598.6%—
ProofBench50%—

Knowledge DeepSeek V4 Pro leads

DeepSeek V4 Pro: 59.5 (#31), Qwen3.5 27B: 38.0 (#150)

BenchmarkDeepSeek V4 ProQwen3.5 27B
Vectara Hallucination Rate8.6%12.1%
LMArena Expert14641428
GPQA Diamond91.7%—
SimpleQA Verified52.9%—

Multimodal Not comparable

DeepSeek V4 Pro: —, Qwen3.5 27B: 39.4 (#59)

BenchmarkDeepSeek V4 ProQwen3.5 27B
LMArena Vision—1241

Multilingual DeepSeek V4 Pro leads

DeepSeek V4 Pro: 54.4 (#45), Qwen3.5 27B: 50.8 (#115)

BenchmarkDeepSeek V4 ProQwen3.5 27B
LMArena Non-English14391390
LMArena Chinese14861478
LMArena French14721410
LMArena German14581393
LMArena Japanese14451345
LMArena Korean14471358
LMArena Russian14531390
LMArena Spanish14581407

Instruction Following DeepSeek V4 Pro leads

DeepSeek V4 Pro: 76.1 (#47), Qwen3.5 27B: 73.5 (#119)

BenchmarkDeepSeek V4 ProQwen3.5 27B
LMArena Instruction Following14481393

Long Context DeepSeek V4 Pro leads

DeepSeek V4 Pro: 45.0 (#51), Qwen3.5 27B: 43.1 (#106)

BenchmarkDeepSeek V4 ProQwen3.5 27B
LMArena Longer Query14581413
CL-bench Life13.5%—

Writing & Preference DeepSeek V4 Pro leads

DeepSeek V4 Pro: 65.5 (#46), Qwen3.5 27B: 59.3 (#111)

BenchmarkDeepSeek V4 ProQwen3.5 27B
LMArena Text14511409
LMArena Creative Writing14461362
LMArena Multi-Turn14671410
EQ-Bench Creative Writing1553—
EQ-Bench 41166—

Frequently asked questions

Is DeepSeek V4 Pro better than Qwen3.5 27B?

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 41.9 on the Noometry Index.

Which is cheaper, DeepSeek V4 Pro or Qwen3.5 27B?

Qwen3.5 27B is cheaper. It lists at $0.30 per million input tokens and $2.40 per million output tokens; DeepSeek V4 Pro lists at $0.66 and $1.98.

Is DeepSeek V4 Pro or Qwen3.5 27B better for coding?

DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 38.9 in the Noometry coding category.

Which has the bigger context window?

DeepSeek V4 Pro does, with 1M tokens against 262K.

How many benchmarks do DeepSeek V4 Pro and Qwen3.5 27B share?

26 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Qwen3.5 27B has 28.

Related comparisons

Go deeper