Noometry

Model comparison

GPT-6 Luna vs Qwen3.5 27B

GPT-6 Luna is the stronger model overall, scoring 53.3 to 41.9 on the Noometry Index.

Last verified . 23 shared benchmarks.

GPT-6 Luna OpenAI

53.3

Rank #36 Confirmed

Qwen3.5 27B Alibaba (Qwen)

41.9

Rank #127 Confirmed

Summary

  • They share 23 benchmarks with published results for both. GPT-6 Luna scores higher in 6 categories and Qwen3.5 27B in 3 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-6 Luna leads 76.1 to 38.8.
  • The biggest single-benchmark swing is NYT Connections (extended): 68.7% for GPT-6 Luna and 47.9% for Qwen3.5 27B.
  • GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.30 / $2.40 for Qwen3.5 27B.
  • GPT-6 Luna accepts more context: 1.05M tokens versus 262K.
  • Qwen3.5 27B has downloadable open weights; the other is API-only.

Side by side

GPT-6 LunaQwen3.5 27B
ProviderOpenAIAlibaba (Qwen)
Noometry Index53.341.9
Released2026-09-222026-02-23
WeightsProprietaryOpen
Context window1.05M262K
Max output128K66K
Input $ / M tokens$0.10$0.30
Output $ / M tokens$0.50$2.40
Results tracked4228

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

Category by category

Coding GPT-6 Luna leads

GPT-6 Luna: 55.5 (#25), Qwen3.5 27B: 38.9 (#168)

BenchmarkGPT-6 LunaQwen3.5 27B
LMArena WebDev15811358
LMArena Coding14391427
ALE-Bench1,577349.45
DeepSWE66.6%—
FrontierCode42.4%—
SciCode54.6%—
WeirdML—39.5%

Agentic & Tool Use Not comparable

GPT-6 Luna: 33.3 (#54), Qwen3.5 27B: —

BenchmarkGPT-6 LunaQwen3.5 27B
APEX-Agents44.3%—
GDP.pdf23%—
Vending-Bench 2—201.98

Reasoning GPT-6 Luna leads

GPT-6 Luna: 48.2 (#41), Qwen3.5 27B: 27.5 (#117)

BenchmarkGPT-6 LunaQwen3.5 27B
NYT Connections (extended)68.7%47.9%
LMArena Hard Prompts14111414
DTBench90.1%82.4%
LMCA44.5%34%
ARC-AGI-259.3%—
ARC-AGI-186.7%—
CritPt19.4%—
Chess Puzzles31%—
Thematic Generalization—45.5%
Mystery Game Puzzles7%—
Epoch Capabilities Index156.28—

Math GPT-6 Luna leads

GPT-6 Luna: 76.1 (#15), Qwen3.5 27B: 38.8 (#127)

BenchmarkGPT-6 LunaQwen3.5 27B
LMArena Math14161429
FrontierMath (Tiers 1-3)78.9%—
FrontierMath Tier 456.1%—
MathArena Final-Answer Competitions—56.7%
OTIS Mock AIME 2024-202598.9%—
ProofBench64%—

Knowledge GPT-6 Luna leads

GPT-6 Luna: 57.0 (#41), Qwen3.5 27B: 38.0 (#150)

BenchmarkGPT-6 LunaQwen3.5 27B
LMArena Expert14441428
GPQA Diamond90.5%—
SimpleQA Verified41.4%—
Vectara Hallucination Rate—12.1%

Multimodal GPT-6 Luna leads

GPT-6 Luna: 42.4 (#30), Qwen3.5 27B: 39.4 (#59)

BenchmarkGPT-6 LunaQwen3.5 27B
LMArena Vision12171241
Blueprint-Bench 231.2%—
Furniture Assembly44.2%—

Multilingual Too close to call

GPT-6 Luna: 50.5 (#117), Qwen3.5 27B: 50.8 (#115)

BenchmarkGPT-6 LunaQwen3.5 27B
LMArena Non-English13861390
LMArena Chinese14331478
LMArena French14201410
LMArena German13691393
LMArena Japanese13691345
LMArena Korean13601358
LMArena Russian13941390
LMArena Spanish13931407

Instruction Following Too close to call

GPT-6 Luna: 74.3 (#99), Qwen3.5 27B: 73.5 (#119)

BenchmarkGPT-6 LunaQwen3.5 27B
LMArena Instruction Following14091393

Long Context Too close to call

GPT-6 Luna: 43.0 (#111), Qwen3.5 27B: 43.1 (#106)

BenchmarkGPT-6 LunaQwen3.5 27B
LMArena Longer Query14091413

Writing & Preference Qwen3.5 27B leads

GPT-6 Luna: 58.3 (#119), Qwen3.5 27B: 59.3 (#111)

BenchmarkGPT-6 LunaQwen3.5 27B
LMArena Text13911409
LMArena Creative Writing13631362
LMArena Multi-Turn13961410

Frequently asked questions

Is GPT-6 Luna better than Qwen3.5 27B?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 41.9 on the Noometry Index.

Which is cheaper, GPT-6 Luna or Qwen3.5 27B?

GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Qwen3.5 27B lists at $0.30 and $2.40.

Is GPT-6 Luna or Qwen3.5 27B better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 38.9 in the Noometry coding category.

Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 262K.

How many benchmarks do GPT-6 Luna and Qwen3.5 27B share?

23 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Qwen3.5 27B has 28.

Related comparisons

Go deeper