Noometry

Model comparison

Qwen2.5 72B Instruct vs Qwen3.5 27B

Qwen3.5 27B is the stronger model overall, scoring 41.9 to 31.9 on the Noometry Index.

Last verified . 20 shared benchmarks.

Qwen2.5 72B Instruct Alibaba (Qwen)

31.9

Rank #267 Confirmed

Qwen3.5 27B Alibaba (Qwen)

41.9

Rank #127 Confirmed

Summary

  • They share 20 benchmarks with published results for both. Qwen2.5 72B Instruct scores higher in 0 categories and Qwen3.5 27B in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Qwen3.5 27B leads 38.8 to 19.3.
  • The biggest single-benchmark swing is WeirdML: 16% for Qwen2.5 72B Instruct and 39.5% for Qwen3.5 27B.
  • Qwen3.5 27B is cheaper at $0.30 / $2.40 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
  • Qwen3.5 27B accepts more context: 262K tokens versus 131K.

Side by side

Qwen2.5 72B InstructQwen3.5 27B
ProviderAlibaba (Qwen)Alibaba (Qwen)
Noometry Index31.941.9
Released2024-092026-02-23
WeightsOpenOpen
Context window131K262K
Max output8K66K
Input $ / M tokens$1.40$0.30
Output $ / M tokens$5.60$2.40
Results tracked4328

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

Category by category

Coding Qwen3.5 27B leads

Qwen2.5 72B Instruct: 33.2 (#260), Qwen3.5 27B: 38.9 (#168)

BenchmarkQwen2.5 72B InstructQwen3.5 27B
WeirdML16%39.5%
LMArena Coding12921427
LMArena WebDev—1358
BigCodeBench Instruct45.8%—
BigCodeBench Complete55.9%—
ALE-Bench—349.45

Agentic & Tool Use Not comparable

Qwen2.5 72B Instruct: 22.1 (#133), Qwen3.5 27B: —

BenchmarkQwen2.5 72B InstructQwen3.5 27B
TheAgentCompany5.7%—
BALROG16.2%—
METR Time Horizons35.8%—
Vending-Bench 2—201.98

Reasoning Qwen3.5 27B leads

Qwen2.5 72B Instruct: 22.3 (#199), Qwen3.5 27B: 27.5 (#117)

BenchmarkQwen2.5 72B InstructQwen3.5 27B
LMArena Hard Prompts12711414
DTBench62.9%82.4%
LMCA13.4%34%
NYT Connections (extended)—47.9%
Thematic Generalization—45.5%
BIG-Bench Hard79.8%—
Epoch Capabilities Index129—
ForecastBench57.5—
HellaSwag84.8%—
PIQA82.6%—
WinoGrande82.3%—

Math Qwen3.5 27B leads

Qwen2.5 72B Instruct: 19.3 (#287), Qwen3.5 27B: 38.8 (#127)

BenchmarkQwen2.5 72B InstructQwen3.5 27B
LMArena Math12831429
MathArena Final-Answer Competitions—56.7%
OTIS Mock AIME 2024-20258.1%—
Omni-MATH33%—
MATH Level 563.2%—

Knowledge Qwen3.5 27B leads

Qwen2.5 72B Instruct: 27.0 (#253), Qwen3.5 27B: 38.0 (#150)

BenchmarkQwen2.5 72B InstructQwen3.5 27B
LMArena Expert12451428
GPQA Diamond49.1%—
MMLU-Pro63.1%—
Confabulations19.1%—
Vectara Hallucination Rate—12.1%
GPQA (HELM)42.6%—
ARC (AI2) Challenge94.5%—
MMLU85.3%—
TriviaQA71.9%—

Multimodal Not comparable

Qwen2.5 72B Instruct: —, Qwen3.5 27B: 39.4 (#59)

BenchmarkQwen2.5 72B InstructQwen3.5 27B
LMArena Vision—1241

Multilingual Qwen3.5 27B leads

Qwen2.5 72B Instruct: 41.0 (#213), Qwen3.5 27B: 50.8 (#115)

BenchmarkQwen2.5 72B InstructQwen3.5 27B
LMArena Non-English12521390
LMArena Chinese12721478
LMArena French12801410
LMArena German12341393
LMArena Japanese11801345
LMArena Korean11881358
LMArena Russian12641390
LMArena Spanish12561407

Instruction Following Qwen3.5 27B leads

Qwen2.5 72B Instruct: 65.5 (#221), Qwen3.5 27B: 73.5 (#119)

BenchmarkQwen2.5 72B InstructQwen3.5 27B
LMArena Instruction Following12541393
IFEval80.6%—

Long Context Qwen3.5 27B leads

Qwen2.5 72B Instruct: 38.9 (#188), Qwen3.5 27B: 43.1 (#106)

BenchmarkQwen2.5 72B InstructQwen3.5 27B
LMArena Longer Query12821413

Writing & Preference Qwen3.5 27B leads

Qwen2.5 72B Instruct: 46.7 (#215), Qwen3.5 27B: 59.3 (#111)

BenchmarkQwen2.5 72B InstructQwen3.5 27B
LMArena Text12691409
LMArena Creative Writing12211362
LMArena Multi-Turn12721410
WildBench80.2%—

Frequently asked questions

Is Qwen2.5 72B Instruct better than Qwen3.5 27B?

Qwen3.5 27B is the stronger model overall, scoring 41.9 to 31.9 on the Noometry Index.

Which is cheaper, Qwen2.5 72B Instruct 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; Qwen2.5 72B Instruct lists at $1.40 and $5.60.

Is Qwen2.5 72B Instruct or Qwen3.5 27B better for coding?

Qwen3.5 27B scores higher on coding benchmarks: 38.9 versus 33.2 in the Noometry coding category.

Which has the bigger context window?

Qwen3.5 27B does, with 262K tokens against 131K.

How many benchmarks do Qwen2.5 72B Instruct and Qwen3.5 27B share?

20 benchmarks have published results for both models. Qwen2.5 72B Instruct has 43 scored results on Noometry and Qwen3.5 27B has 28.

Related comparisons

Go deeper