Noometry

Model comparison

Grok-2 (Dec 2024) vs Qwen3.8 27B

Qwen3.8 27B is the stronger model overall, scoring 46.0 to 33.7 on the Noometry Index.

Last verified . 19 shared benchmarks.

Grok-2 (Dec 2024) xAI

33.7

Rank #239 Confirmed

Qwen3.8 27B Alibaba (Qwen)

46.0

Rank #68 Confirmed

Summary

  • They share 19 benchmarks with published results for both. Grok-2 (Dec 2024) scores higher in 0 categories and Qwen3.8 27B in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Qwen3.8 27B leads 41.0 to 16.9.
  • The biggest single-benchmark swing is DTBench: 65.2% for Grok-2 (Dec 2024) and 88% for Qwen3.8 27B.
  • Qwen3.8 27B has downloadable open weights; the other is API-only.

Side by side

Grok-2 (Dec 2024)Qwen3.8 27B
ProviderxAIAlibaba (Qwen)
Noometry Index33.746.0
Released2024-08-132026-08-14
WeightsProprietaryOpen
Context window—262K
Max output—33K
Input $ / M tokens—$0.99
Output $ / M tokens—$1.49
Results tracked3431

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

Category by category

Coding Qwen3.8 27B leads

Grok-2 (Dec 2024): 33.3 (#258), Qwen3.8 27B: 50.5 (#44)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
LMArena Coding12871482
LMArena WebDev—1593
SciCode—46.6%
WeirdML22.2%—
LiveBench Coding46.4%—

Agentic & Tool Use Not comparable

Grok-2 (Dec 2024): —, Qwen3.8 27B: 32.9 (#57)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
APEX-Agents—47.5%

Reasoning Qwen3.8 27B leads

Grok-2 (Dec 2024): 16.9 (#299), Qwen3.8 27B: 41.0 (#54)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
LMArena Hard Prompts12721460
DTBench65.2%88%
Epoch Capabilities Index130.48149.38
ARC-AGI-2—42.4%
SimpleBench22.7%—
NYT Connections (extended)—54.5%
ARC-AGI-1—87.5%
CritPt—5.4%
LiveBench Reasoning54.8%—
LiveBench Data Analysis54.5%—
LMCA—41.4%
Surface Evolver Bench—45%
LiveBench54.3%—

Math Qwen3.8 27B leads

Grok-2 (Dec 2024): 20.8 (#284), Qwen3.8 27B: 37.1 (#161)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
LMArena Math12831456
OTIS Mock AIME 2024-202511.5%—
ProofBench—16%
LiveBench Math54.9%—
MATH Level 563.5%—
FrontierMath (Feb 2025 set)0.7%—

Knowledge Qwen3.8 27B leads

Grok-2 (Dec 2024): 29.8 (#233), Qwen3.8 27B: 41.6 (#109)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
LMArena Expert12541482
GPQA Diamond53.8%—
Confabulations20.1%—

Multimodal Not comparable

Grok-2 (Dec 2024): —, Qwen3.8 27B: 41.3 (#37)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
LMArena Vision—1271

Multilingual Qwen3.8 27B leads

Grok-2 (Dec 2024): 43.1 (#188), Qwen3.8 27B: 53.7 (#60)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
LMArena Non-English12821430
LMArena Chinese12891504
LMArena French13181465
LMArena German12871438
LMArena Japanese12441384
LMArena Korean12371393
LMArena Russian12861415
LMArena Spanish12811448

Instruction Following Qwen3.8 27B leads

Grok-2 (Dec 2024): 66.9 (#202), Qwen3.8 27B: 75.8 (#53)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
LMArena Instruction Following12701439
LiveBench Instruction Following69.6%—

Long Context Qwen3.8 27B leads

Grok-2 (Dec 2024): 38.8 (#190), Qwen3.8 27B: 44.3 (#70)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
LMArena Longer Query12761450

Writing & Preference Qwen3.8 27B leads

Grok-2 (Dec 2024): 48.6 (#198), Qwen3.8 27B: 65.8 (#43)

BenchmarkGrok-2 (Dec 2024)Qwen3.8 27B
LMArena Text13051441
LMArena Creative Writing12841384
LMArena Multi-Turn12901441
Short-Story Creative Writing63.6%—
EQ-Bench Creative Writing—1671
LiveBench Language45.6%—

Frequently asked questions

Is Grok-2 (Dec 2024) better than Qwen3.8 27B?

Qwen3.8 27B is the stronger model overall, scoring 46.0 to 33.7 on the Noometry Index.

Is Grok-2 (Dec 2024) or Qwen3.8 27B better for coding?

Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 33.3 in the Noometry coding category.

How many benchmarks do Grok-2 (Dec 2024) and Qwen3.8 27B share?

19 benchmarks have published results for both models. Grok-2 (Dec 2024) has 34 scored results on Noometry and Qwen3.8 27B has 31.

Related comparisons

Go deeper