Noometry

Model comparison

Grok-3 mini vs Qwen3-Next 80B-A3B Instruct

Qwen3-Next 80B-A3B Instruct is the stronger model overall, scoring 43.0 to 41.2 on the Noometry Index.

Last verified . 24 shared benchmarks.

Grok-3 mini xAI

41.2

Rank #141 Confirmed

Qwen3-Next 80B-A3B Instruct Alibaba (Qwen)

43.0

Rank #102 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Grok-3 mini scores higher in 4 categories and Qwen3-Next 80B-A3B Instruct in 4 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Qwen3-Next 80B-A3B Instruct leads 31.1 to 13.6.
  • The biggest single-benchmark swing is WildBench: 65.1% for Grok-3 mini and 80.7% for Qwen3-Next 80B-A3B Instruct.
  • Qwen3-Next 80B-A3B Instruct has downloadable open weights; the other is API-only.

Side by side

Grok-3 miniQwen3-Next 80B-A3B Instruct
ProviderxAIAlibaba (Qwen)
Noometry Index41.243.0
Released2025-04-092025-09
WeightsProprietaryOpen
Context window—131K
Max output—33K
Input $ / M tokens—$0.50
Output $ / M tokens—$2
Results tracked3525

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

Category by category

Coding Qwen3-Next 80B-A3B Instruct leads

Grok-3 mini: 40.8 (#131), Qwen3-Next 80B-A3B Instruct: 42.5 (#98)

BenchmarkGrok-3 miniQwen3-Next 80B-A3B Instruct
LMArena Coding13791440
Aider Polyglot49.3%—
WeirdML42.6%—

Reasoning Qwen3-Next 80B-A3B Instruct leads

Grok-3 mini: 13.6 (#334), Qwen3-Next 80B-A3B Instruct: 31.1 (#81)

BenchmarkGrok-3 miniQwen3-Next 80B-A3B Instruct
Kagi LLM Benchmark61.3%66.7%
LMArena Hard Prompts13751428
ARC-AGI-20.4%—
ARC-AGI-116.5%—
Epoch Capabilities Index140.35—

Math Grok-3 mini leads

Grok-3 mini: 42.1 (#85), Qwen3-Next 80B-A3B Instruct: 38.8 (#126)

BenchmarkGrok-3 miniQwen3-Next 80B-A3B Instruct
Omni-MATH31.8%46.7%
LMArena Math13861440
OTIS Mock AIME 2024-202577.8%—
MATH Level 590.9%—
FrontierMath (Feb 2025 set)5.9%—

Knowledge Grok-3 mini leads

Grok-3 mini: 46.4 (#81), Qwen3-Next 80B-A3B Instruct: 41.8 (#106)

BenchmarkGrok-3 miniQwen3-Next 80B-A3B Instruct
MMLU-Pro79.9%78.6%
GPQA (HELM)67.5%63%
LMArena Expert13951417
GPQA Diamond76.3%—
Confabulations10.8%—
Vectara Hallucination Rate—9.3%

Multilingual Qwen3-Next 80B-A3B Instruct leads

Grok-3 mini: 48.1 (#145), Qwen3-Next 80B-A3B Instruct: 52.1 (#93)

BenchmarkGrok-3 miniQwen3-Next 80B-A3B Instruct
LMArena Non-English13521407
LMArena Chinese13871460
LMArena French13571413
LMArena German13491417
LMArena Japanese13421395
LMArena Korean13351364
LMArena Russian13531404
LMArena Spanish13811435

Instruction Following Grok-3 mini leads

Grok-3 mini: 78.5 (#9), Qwen3-Next 80B-A3B Instruct: 70.8 (#159)

BenchmarkGrok-3 miniQwen3-Next 80B-A3B Instruct
IFEval95.1%81%
LMArena Instruction Following13571389

Long Context Grok-3 mini leads

Grok-3 mini: 41.0 (#147), Qwen3-Next 80B-A3B Instruct: 37.0 (#223)

BenchmarkGrok-3 miniQwen3-Next 80B-A3B Instruct
Fiction.LiveBench66.7%55.6%
LMArena Longer Query13721403

Writing & Preference Qwen3-Next 80B-A3B Instruct leads

Grok-3 mini: 52.5 (#169), Qwen3-Next 80B-A3B Instruct: 58.0 (#121)

BenchmarkGrok-3 miniQwen3-Next 80B-A3B Instruct
LMArena Text13701417
LMArena Creative Writing13421334
WildBench65.1%80.7%
LMArena Multi-Turn13551416
Short-Story Creative Writing73.5%—

Frequently asked questions

Is Grok-3 mini better than Qwen3-Next 80B-A3B Instruct?

Qwen3-Next 80B-A3B Instruct is the stronger model overall, scoring 43.0 to 41.2 on the Noometry Index.

Is Grok-3 mini or Qwen3-Next 80B-A3B Instruct better for coding?

Qwen3-Next 80B-A3B Instruct scores higher on coding benchmarks: 42.5 versus 40.8 in the Noometry coding category.

How many benchmarks do Grok-3 mini and Qwen3-Next 80B-A3B Instruct share?

24 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Qwen3-Next 80B-A3B Instruct has 25.

Related comparisons

Go deeper