Noometry

Model comparison

GPT-5.6 Luna vs Hy3

GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 44.2 on the Noometry Index. Hy3 costs 2.0× less per token, which makes it the better buy when GPT-5.6 Luna's lead doesn't matter for your workload.

Last verified . 19 shared benchmarks.

GPT-5.6 Luna OpenAI

54.6

Rank #30 Confirmed

Hy3 Tencent

44.2

Rank #79 Confirmed

Summary

  • They share 19 benchmarks with published results for both. GPT-5.6 Luna scores higher in 6 categories and Hy3 in 2 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 40.1.
  • The biggest single-benchmark swing is NYT Connections (extended): 69.4% for GPT-5.6 Luna and 41.2% for Hy3.
  • Hy3 is cheaper at $0.13 / $0.53 per million input/output tokens, against $0.20 / $1.20 for GPT-5.6 Luna.
  • GPT-5.6 Luna accepts more context: 1.05M tokens versus 262K.
  • Hy3 has downloadable open weights; the other is API-only.

Side by side

GPT-5.6 LunaHy3
ProviderOpenAITencent
Noometry Index54.644.2
Released2026-07-092026-07-06
WeightsProprietaryOpen
Context window1.05M262K
Max output128K128K
Input $ / M tokens$0.20$0.13
Output $ / M tokens$1.20$0.53
Results tracked5219

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

Category by category

Coding GPT-5.6 Luna leads

GPT-5.6 Luna: 54.5 (#28), Hy3: 46.8 (#63)

BenchmarkGPT-5.6 LunaHy3
LMArena WebDev15191508
LMArena Coding14661464
DeepSWE67.2%—
FrontierCode39.8%—
CursorBench35.9%—
SciCode53.6%—
WeirdML60.9%—
ALE-Bench1,667—

Agentic & Tool Use Not comparable

GPT-5.6 Luna: 34.4 (#45), Hy3: —

BenchmarkGPT-5.6 LunaHy3
APEX-Agents43%—
BALROG45.6%—
GDP.pdf22.7%—
Vending-Bench 24,095—

Reasoning GPT-5.6 Luna leads

GPT-5.6 Luna: 47.6 (#43), Hy3: 26.1 (#136)

BenchmarkGPT-5.6 LunaHy3
NYT Connections (extended)69.4%41.2%
LMArena Hard Prompts14511447
ARC-AGI-259.5%—
SimpleBench46.8%—
Kagi LLM Benchmark49.1%—
ARC-AGI-188%—
CritPt20.6%—
Chess Puzzles40%—
Mystery Game Puzzles21%—
DTBench89.1%—
LMCA48.5%—
Surface Evolver Bench61.9%—
Epoch Capabilities Index156.39—

Math GPT-5.6 Luna leads

GPT-5.6 Luna: 77.7 (#14), Hy3: 40.1 (#93)

BenchmarkGPT-5.6 LunaHy3
LMArena Math14581475
FrontierMath (Tiers 1-3)82.1%—
FrontierMath Tier 461%—
OTIS Mock AIME 2024-202598.3%—
ProofBench60%—

Knowledge GPT-5.6 Luna leads

GPT-5.6 Luna: 58.5 (#34), Hy3: 40.8 (#114)

BenchmarkGPT-5.6 LunaHy3
LMArena Expert14781460
GPQA Diamond91.6%—
SimpleQA Verified41%—

Multimodal Not comparable

GPT-5.6 Luna: 42.7 (#28), Hy3: —

BenchmarkGPT-5.6 LunaHy3
LMArena Vision1258—
Blueprint-Bench 222.6%—
Furniture Assembly42.5%—
LMArena Document1457—

Multilingual Too close to call

GPT-5.6 Luna: 52.8 (#78), Hy3: 53.5 (#65)

BenchmarkGPT-5.6 LunaHy3
LMArena Non-English14171426
LMArena Chinese14701493
LMArena French14561461
LMArena German14541439
LMArena Japanese14111392
LMArena Korean14151395
LMArena Russian14281432
LMArena Spanish14481456

Instruction Following Too close to call

GPT-5.6 Luna: 75.6 (#57), Hy3: 75.1 (#70)

BenchmarkGPT-5.6 LunaHy3
LMArena Instruction Following14371426

Long Context Too close to call

GPT-5.6 Luna: 43.9 (#82), Hy3: 44.1 (#75)

BenchmarkGPT-5.6 LunaHy3
LMArena Longer Query14361442

Writing & Preference GPT-5.6 Luna leads

GPT-5.6 Luna: 68.0 (#29), Hy3: 62.2 (#81)

BenchmarkGPT-5.6 LunaHy3
LMArena Text14311439
LMArena Creative Writing13961402
LMArena Multi-Turn14341436
EQ-Bench Creative Writing1829—
EQ-Bench 41156—

Frequently asked questions

Is GPT-5.6 Luna better than Hy3?

GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 44.2 on the Noometry Index. Hy3 costs 2.0× less per token, which makes it the better buy when GPT-5.6 Luna's lead doesn't matter for your workload.

Which is cheaper, GPT-5.6 Luna or Hy3?

Hy3 is cheaper. It lists at $0.13 per million input tokens and $0.53 per million output tokens; GPT-5.6 Luna lists at $0.20 and $1.20.

Is GPT-5.6 Luna or Hy3 better for coding?

GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 46.8 in the Noometry coding category.

Which has the bigger context window?

GPT-5.6 Luna does, with 1.05M tokens against 262K.

How many benchmarks do GPT-5.6 Luna and Hy3 share?

19 benchmarks have published results for both models. GPT-5.6 Luna has 52 scored results on Noometry and Hy3 has 19.

Related comparisons

Go deeper