Noometry

Model comparison

Grok 4.20 (Non-Reasoning) vs Qwen2.5-Coder-32B

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 32.9 on the Noometry Index. Qwen2.5-Coder-32B costs 2.1× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.

Last verified . 13 shared benchmarks.

Grok 4.20 (Non-Reasoning) xAI

48.5

Rank #48 Confirmed

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

32.9

Rank #239 Confirmed

Summary

  • They share 13 benchmarks with published results for both. Grok 4.20 (Non-Reasoning) scores higher in 8 categories and Qwen2.5-Coder-32B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.2 to 20.4.
  • Qwen2.5-Coder-32B is cheaper at $0.66 / $1 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 (Non-Reasoning).
  • Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 33K.
  • Qwen2.5-Coder-32B has downloadable open weights; the other is API-only.

Side by side

Grok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
ProviderxAIAlibaba (Qwen)
Noometry Index48.532.9
Released2026-02-172024-09-18
WeightsProprietaryOpen
Context window1M33K
Max output30K29K
Input $ / M tokens$1.25$0.66
Output $ / M tokens$2.50$1
Results tracked4631

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

Category by category

Coding Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 41.8 (#104), Qwen2.5-Coder-32B: 22.4 (#328)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
LMArena Coding14581276
SWE-bench Verified (bash only)—9%
Aider Polyglot—16.4%
LMArena WebDev1374—
WeirdML52.3%—
BigCodeBench Instruct—49%
LiveBench Coding—56.9%
BigCodeBench Complete—58%
ALE-Bench1,150—
HumanEval+—87.2%
MBPP+—77%

Agentic & Tool Use Not comparable

Grok 4.20 (Non-Reasoning): 34.3 (#43), Qwen2.5-Coder-32B: —

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
Terminal-Bench57.3%—
τ²-bench Banking18%—
LMArena Search1189—
Vending-Bench 24,663—

Reasoning Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.2 (#30), Qwen2.5-Coder-32B: 20.4 (#220)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
LMArena Hard Prompts14511251
Epoch Capabilities Index152.01119.41
ARC-AGI-265.1%—
Kagi LLM Benchmark75%—
NYT Connections (extended)85.4%—
ARC-AGI-189.5%—
Chess Puzzles24%—
Thematic Generalization63.8%—
LiveBench Reasoning—42.1%
DTBench90.1%—
LiveBench Data Analysis—49.9%
LMCA38.7%—
ForecastBench61.4—
HellaSwag—83%
LiveBench—46.2%
WinoGrande—80.8%

Math Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 48.0 (#54), Qwen2.5-Coder-32B: 30.8 (#203)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
LMArena Math14551251
FrontierMath (Tiers 1-3)44.9%—
FrontierMath Tier 417.1%—
OTIS Mock AIME 2024-202592.2%—
ProofBench14%—
LiveBench Math—46.6%
GSM8K—93%

Knowledge Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.8 (#54), Qwen2.5-Coder-32B: 32.9 (#200)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
LMArena Expert14401221
GPQA Diamond89.3%—
SimpleQA Verified30.2%—
ARC (AI2) Challenge—70.5%
MMLU—79.1%

Multimodal Not comparable

Grok 4.20 (Non-Reasoning): 33.2 (#88), Qwen2.5-Coder-32B: —

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
LMArena Vision1263—
Blueprint-Bench 20%—
LMArena Document1416—

Multilingual Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 54.7 (#38), Qwen2.5-Coder-32B: 38.0 (#230)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
LMArena Non-English14401205
LMArena Chinese14811222
LMArena Russian14571228
LMArena French1477—
LMArena German1463—
LMArena Japanese1450—
LMArena Korean1417—
LMArena Spanish1442—

Instruction Following Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 75.0 (#79), Qwen2.5-Coder-32B: 61.6 (#240)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
LMArena Instruction Following14201223
LiveBench Instruction Following—58.7%

Long Context Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 45.4 (#33), Qwen2.5-Coder-32B: 37.9 (#204)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
LMArena Longer Query14361251
CL-bench22.2%—
CL-bench Life11.9%—

Writing & Preference Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 65.9 (#42), Qwen2.5-Coder-32B: 41.8 (#235)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen2.5-Coder-32B
LMArena Text14511230
LMArena Creative Writing14391174
LMArena Multi-Turn14561222
EQ-Bench Creative Writing1574—
LiveBench Language—23.3%

Frequently asked questions

Is Grok 4.20 (Non-Reasoning) better than Qwen2.5-Coder-32B?

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 32.9 on the Noometry Index. Qwen2.5-Coder-32B costs 2.1× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.

Which is cheaper, Grok 4.20 (Non-Reasoning) or Qwen2.5-Coder-32B?

Qwen2.5-Coder-32B is cheaper. It lists at $0.66 per million input tokens and $1 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.

Is Grok 4.20 (Non-Reasoning) or Qwen2.5-Coder-32B better for coding?

Grok 4.20 (Non-Reasoning) scores higher on coding benchmarks: 41.8 versus 22.4 in the Noometry coding category.

Which has the bigger context window?

Grok 4.20 (Non-Reasoning) does, with 1M tokens against 33K.

How many benchmarks do Grok 4.20 (Non-Reasoning) and Qwen2.5-Coder-32B share?

13 benchmarks have published results for both models. Grok 4.20 (Non-Reasoning) has 46 scored results on Noometry and Qwen2.5-Coder-32B has 31.

Related comparisons

Go deeper