Noometry

Model comparison

Qwen2.5-Coder-32B vs Qwen2.5-VL 72B Instruct

Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 29.4 on the Noometry Index.

Last verified . 0 shared benchmarks.

Qwen2.5-Coder-32B Alibaba (Qwen)

32.9

Rank #239 Confirmed

Qwen2.5-VL 72B Instruct Alibaba (Qwen)

29.4

Rank #298 Confirmed

Summary

  • Qwen2.5-Coder-32B is cheaper at $0.66 / $1 per million input/output tokens, against $2.80 / $8.40 for Qwen2.5-VL 72B Instruct.
  • Qwen2.5-VL 72B Instruct accepts more context: 131K tokens versus 33K.

Side by side

Qwen2.5-Coder-32BQwen2.5-VL 72B Instruct
ProviderAlibaba (Qwen)Alibaba (Qwen)
Noometry Index32.929.4
Released2024-09-182024-09
WeightsOpenOpen
Context window33K131K
Max output29K8K
Input $ / M tokens$0.66$2.80
Output $ / M tokens$1$8.40
Results tracked316

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

Category by category

Coding Not comparable

Qwen2.5-Coder-32B: 22.4 (#328), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
SWE-bench Verified (bash only)9%—
Aider Polyglot16.4%—
BigCodeBench Instruct49%—
LiveBench Coding56.9%—
LMArena Coding1276—
BigCodeBench Complete58%—
HumanEval+87.2%—
MBPP+77%—

Agentic & Tool Use Not comparable

Qwen2.5-Coder-32B: —, Qwen2.5-VL 72B Instruct: 18.4 (#136)

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
OSWorld—5%

Reasoning Too close to call

Qwen2.5-Coder-32B: 20.4 (#220), Qwen2.5-VL 72B Instruct: 20.0 (#228)

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
Kagi LLM Benchmark—36%
LiveBench Reasoning42.1%—
LMArena Hard Prompts1251—
LiveBench Data Analysis49.9%—
Epoch Capabilities Index119.41—
HellaSwag83%—
LiveBench46.2%—
WinoGrande80.8%—

Math Not comparable

Qwen2.5-Coder-32B: 30.8 (#203), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
LiveBench Math46.6%—
LMArena Math1251—
GSM8K93%—

Knowledge Not comparable

Qwen2.5-Coder-32B: 32.9 (#200), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
LMArena Expert1221—
ARC (AI2) Challenge70.5%—
MMLU79.1%—

Multimodal Not comparable

Qwen2.5-Coder-32B: —, Qwen2.5-VL 72B Instruct: 32.9 (#92)

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
LMArena Vision—1107
Video-MME—73.5%
GeoBench—62%
SpatialViz-Bench—33.3%

Multilingual Not comparable

Qwen2.5-Coder-32B: 38.0 (#230), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
LMArena Non-English1205—
LMArena Chinese1222—
LMArena Russian1228—

Instruction Following Not comparable

Qwen2.5-Coder-32B: 61.6 (#240), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
LiveBench Instruction Following58.7%—
LMArena Instruction Following1223—

Long Context Not comparable

Qwen2.5-Coder-32B: 37.9 (#204), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
LMArena Longer Query1251—

Writing & Preference Not comparable

Qwen2.5-Coder-32B: 41.8 (#235), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen2.5-Coder-32BQwen2.5-VL 72B Instruct
LMArena Text1230—
LMArena Creative Writing1174—
LMArena Multi-Turn1222—
LiveBench Language23.3%—

Frequently asked questions

Is Qwen2.5-Coder-32B better than Qwen2.5-VL 72B Instruct?

Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 29.4 on the Noometry Index.

Which is cheaper, Qwen2.5-Coder-32B or Qwen2.5-VL 72B Instruct?

Qwen2.5-Coder-32B is cheaper. It lists at $0.66 per million input tokens and $1 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.

Which has the bigger context window?

Qwen2.5-VL 72B Instruct does, with 131K tokens against 33K.

How many benchmarks do Qwen2.5-Coder-32B and Qwen2.5-VL 72B Instruct share?

0 benchmarks have published results for both models. Qwen2.5-Coder-32B has 31 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.

Related comparisons

Go deeper