Noometry

Model comparison

GLM-4.6 vs Qwen3.8 27B

Qwen3.8 27B is the stronger model overall, scoring 46.0 to 41.4 on the Noometry Index.

Last verified . 21 shared benchmarks.

GLM-4.6 Z.ai (Zhipu)

41.4

Rank #135 Confirmed

Qwen3.8 27B Alibaba (Qwen)

46.0

Rank #68 Confirmed

Summary

  • They share 21 benchmarks with published results for both. GLM-4.6 scores higher in 1 category and Qwen3.8 27B in 8 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Qwen3.8 27B leads 41.0 to 23.7.
  • The biggest single-benchmark swing is SciCode: 38.4% for GLM-4.6 and 46.6% for Qwen3.8 27B.
  • GLM-4.6 is cheaper at $0.60 / $2.20 per million input/output tokens, against $0.99 / $1.49 for Qwen3.8 27B.
  • Qwen3.8 27B accepts more context: 262K tokens versus 205K.

Side by side

GLM-4.6Qwen3.8 27B
ProviderZ.ai (Zhipu)Alibaba (Qwen)
Noometry Index41.446.0
Released2025-09-302026-08-14
WeightsOpenOpen
Context window205K262K
Max output131K33K
Input $ / M tokens$0.60$0.99
Output $ / M tokens$2.20$1.49
Results tracked2931

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

Category by category

Coding Qwen3.8 27B leads

GLM-4.6: 40.1 (#148), Qwen3.8 27B: 50.5 (#44)

BenchmarkGLM-4.6Qwen3.8 27B
LMArena WebDev13401593
SciCode38.4%46.6%
LMArena Coding14491482
SWE-bench Verified (bash only)55.4%—
ALE-Bench340.82—

Agentic & Tool Use Too close to call

GLM-4.6: 32.3 (#66), Qwen3.8 27B: 32.9 (#57)

BenchmarkGLM-4.6Qwen3.8 27B
Terminal-Bench24.5%—
APEX-Agents—47.5%
Berkeley Function Calling Leaderboard72.4%—

Reasoning Qwen3.8 27B leads

GLM-4.6: 23.7 (#172), Qwen3.8 27B: 41.0 (#54)

BenchmarkGLM-4.6Qwen3.8 27B
CritPt1.1%5.4%
LMArena Hard Prompts14401460
ARC-AGI-2—42.4%
Kagi LLM Benchmark47.4%—
NYT Connections (extended)—54.5%
ARC-AGI-1—87.5%
DTBench—88%
LMCA—41.4%
Surface Evolver Bench—45%
Epoch Capabilities Index—149.38

Math GLM-4.6 leads

GLM-4.6: 39.1 (#111), Qwen3.8 27B: 37.1 (#161)

BenchmarkGLM-4.6Qwen3.8 27B
LMArena Math14321456
ProofBench—16%
FrontierMath (Feb 2025 set)3.8%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Qwen3.8 27B leads

GLM-4.6: 40.2 (#124), Qwen3.8 27B: 41.6 (#109)

BenchmarkGLM-4.6Qwen3.8 27B
LMArena Expert14311482
Vectara Hallucination Rate9.5%—

Multimodal Not comparable

GLM-4.6: —, Qwen3.8 27B: 41.3 (#37)

BenchmarkGLM-4.6Qwen3.8 27B
LMArena Vision—1271

Multilingual Too close to call

GLM-4.6: 53.5 (#66), Qwen3.8 27B: 53.7 (#60)

BenchmarkGLM-4.6Qwen3.8 27B
LMArena Non-English14261430
LMArena Chinese14991504
LMArena French14591465
LMArena German14471438
LMArena Japanese13931384
LMArena Korean14001393
LMArena Russian14191415
LMArena Spanish14361448

Instruction Following Qwen3.8 27B leads

GLM-4.6: 74.3 (#98), Qwen3.8 27B: 75.8 (#53)

BenchmarkGLM-4.6Qwen3.8 27B
LMArena Instruction Following14101439

Long Context Too close to call

GLM-4.6: 43.4 (#94), Qwen3.8 27B: 44.3 (#70)

BenchmarkGLM-4.6Qwen3.8 27B
LMArena Longer Query14221450

Writing & Preference Qwen3.8 27B leads

GLM-4.6: 61.1 (#90), Qwen3.8 27B: 65.8 (#43)

BenchmarkGLM-4.6Qwen3.8 27B
LMArena Text14401441
LMArena Creative Writing14111384
EQ-Bench Creative Writing14111671
LMArena Multi-Turn14271441

Frequently asked questions

Is GLM-4.6 better than Qwen3.8 27B?

Qwen3.8 27B is the stronger model overall, scoring 46.0 to 41.4 on the Noometry Index.

Which is cheaper, GLM-4.6 or Qwen3.8 27B?

GLM-4.6 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; Qwen3.8 27B lists at $0.99 and $1.49.

Is GLM-4.6 or Qwen3.8 27B better for coding?

Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 40.1 in the Noometry coding category.

Which has the bigger context window?

Qwen3.8 27B does, with 262K tokens against 205K.

How many benchmarks do GLM-4.6 and Qwen3.8 27B share?

21 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Qwen3.8 27B has 31.

Related comparisons

Go deeper