Noometry

Model comparison

GPT-6 Luna vs Qwen3.6 Flash

GPT-6 Luna is the stronger model overall, scoring 53.3 to 38.8 on the Noometry Index.

Last verified . 10 shared benchmarks.

GPT-6 Luna OpenAI

53.3

Rank #36 Confirmed

Qwen3.6 Flash Alibaba (Qwen)

38.8

Rank #182 Confirmed

Summary

  • They share 10 benchmarks with published results for both. GPT-6 Luna scores higher in 3 categories and Qwen3.6 Flash in 0 categories; 3 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-6 Luna leads 76.1 to 39.0.
  • The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 78.9% for GPT-6 Luna and 22.5% for Qwen3.6 Flash.
  • GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.19 / $1.13 for Qwen3.6 Flash.
  • GPT-6 Luna accepts more context: 1.05M tokens versus 1M.

Side by side

GPT-6 LunaQwen3.6 Flash
ProviderOpenAIAlibaba (Qwen)
Noometry Index53.338.8
Released2026-09-222026-04-27
WeightsProprietaryProprietary
Context window1.05M1M
Max output128K66K
Input $ / M tokens$0.10$0.19
Output $ / M tokens$0.50$1.13
Results tracked4213

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

Category by category

Coding Not comparable

GPT-6 Luna: 55.5 (#25), Qwen3.6 Flash: —

BenchmarkGPT-6 LunaQwen3.6 Flash
ALE-Bench1,577326.4
DeepSWE66.6%—
FrontierCode42.4%—
LMArena WebDev1581—
SciCode54.6%—
LMArena Coding1439—

Agentic & Tool Use Not comparable

GPT-6 Luna: 33.3 (#54), Qwen3.6 Flash: —

BenchmarkGPT-6 LunaQwen3.6 Flash
APEX-Agents44.3%—
GDP.pdf23%—

Reasoning GPT-6 Luna leads

GPT-6 Luna: 48.2 (#41), Qwen3.6 Flash: 29.0 (#96)

BenchmarkGPT-6 LunaQwen3.6 Flash
Chess Puzzles31%20%
Mystery Game Puzzles7%18%
DTBench90.1%77.1%
LMCA44.5%31%
Epoch Capabilities Index156.28143.26
ARC-AGI-259.3%—
SimpleBench—35.2%
NYT Connections (extended)68.7%—
ARC-AGI-186.7%—
CritPt19.4%—
LMArena Hard Prompts1411—

Math GPT-6 Luna leads

GPT-6 Luna: 76.1 (#15), Qwen3.6 Flash: 39.0 (#117)

BenchmarkGPT-6 LunaQwen3.6 Flash
FrontierMath (Tiers 1-3)78.9%22.5%
OTIS Mock AIME 2024-202598.9%84.4%
FrontierMath Tier 456.1%—
ProofBench64%—
LMArena Math1416—
FrontierMath (Feb 2025 set)—10.3%
FrontierMath Tier 4 (v1)—0%

Knowledge GPT-6 Luna leads

GPT-6 Luna: 57.0 (#41), Qwen3.6 Flash: 42.1 (#100)

BenchmarkGPT-6 LunaQwen3.6 Flash
GPQA Diamond90.5%83.3%
SimpleQA Verified41.4%15.9%
LMArena Expert1444—

Multimodal Not comparable

GPT-6 Luna: 42.4 (#30), Qwen3.6 Flash: —

BenchmarkGPT-6 LunaQwen3.6 Flash
LMArena Vision1217—
Blueprint-Bench 231.2%—
Furniture Assembly44.2%—

Multilingual Not comparable

GPT-6 Luna: 50.5 (#117), Qwen3.6 Flash: —

BenchmarkGPT-6 LunaQwen3.6 Flash
LMArena Non-English1386—
LMArena Chinese1433—
LMArena French1420—
LMArena German1369—
LMArena Japanese1369—
LMArena Korean1360—
LMArena Russian1394—
LMArena Spanish1393—

Instruction Following Not comparable

GPT-6 Luna: 74.3 (#99), Qwen3.6 Flash: —

BenchmarkGPT-6 LunaQwen3.6 Flash
LMArena Instruction Following1409—

Long Context Not comparable

GPT-6 Luna: 43.0 (#111), Qwen3.6 Flash: —

BenchmarkGPT-6 LunaQwen3.6 Flash
LMArena Longer Query1409—

Writing & Preference Not comparable

GPT-6 Luna: 58.3 (#119), Qwen3.6 Flash: —

BenchmarkGPT-6 LunaQwen3.6 Flash
LMArena Text1391—
LMArena Creative Writing1363—
LMArena Multi-Turn1396—

Frequently asked questions

Is GPT-6 Luna better than Qwen3.6 Flash?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 38.8 on the Noometry Index.

Which is cheaper, GPT-6 Luna or Qwen3.6 Flash?

GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Qwen3.6 Flash lists at $0.19 and $1.13.

Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 1M.

How many benchmarks do GPT-6 Luna and Qwen3.6 Flash share?

10 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Qwen3.6 Flash has 13.

Related comparisons

Go deeper