Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

GPT-5.5 vs GPT-6 Luna

24 SHARED BENCHMARKS

Across 24 shared benchmarks, GPT-5.5 scores higher on 20 and GPT-6 Luna on 4. The widest gap is ARC-AGI-3, where GPT-5.5 scores 0.4 against 0.1. GPT-6 Luna is the cheaper of the two on tracked API pricing ($0.10 against $5.00 per million input tokens).

OPENAIVSOPENAI24 SHARED20–4 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerOpenAIOpenAI
Released
Price per 1M tokens input / output$5.00 / $30.00$0.10 / $0.50
Cost of 1M in + 1M out$35.00$0.60 58.3× less
Head-to-head of 24 shared benchmarks20 wins4 wins
Scores tracked independently verified183 30 ◆102 33 ◆

Release dates per Artificial Analysis. Prices: Artificial Analysis for GPT-5.5; Artificial Analysis for GPT-6 Luna. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaGPT-5.5WINSGPT-6 Luna
Reasoning50GPT-5.5 leads 5 of 5 · widest: ARC-AGI-2 85 vs 59.3
Coding31GPT-5.5 leads 3 of 4 · widest: LiveBench · Coding 82.2 vs 79
Agentic11Even, 1–1 of 2
Factuality11Even, 1–1 of 2
Instruction Following10GPT-5.5 leads 1 of 1 · widest: LiveBench · Instruction Following 70.7 vs 55.9
Long Context10GPT-5.5 leads 1 of 1
Math10GPT-5.5 leads 1 of 1 · widest: LiveBench · Mathematics 95.9 vs 89.1
Multimodal10GPT-5.5 leads 1 of 1 · widest: MMMU-Pro 79.9 vs 75.5

Biggest gaps

GPT-5.5 pulls furthest ahead on

  1. ARC-AGI-285 vs 59.3
  2. CritPt27.1 vs 19.4
  3. AA-Omniscience · Accuracy58 vs 43.8

GPT-6 Luna pulls furthest ahead on

  1. AA-Omniscience · Non-hallucination23.3 vs 11
  2. LMArena · WebDev1593 vs 1512

Every shared benchmark24 · grouped by area

Reasoning 5

BenchmarkGPT-5.5MARGINGPT-6 Luna
ARC-AGI-30.4 ◆◆ 0.1
ARC-AGI-285◆ 59.3
CritPt27.119.4
LiveBench · Reasoning89.7 ◆◆ 81.8

Coding 4

BenchmarkGPT-5.5MARGINGPT-6 Luna
LMArena · WebDev1512 ◆◆ 1593
LiveBench · Coding82.2 ◆◆ 79
SciCode55.854.6

Agentic 2

BenchmarkGPT-5.5MARGINGPT-6 Luna
GDPVal41.843.4
Terminal-Bench 4.014.612.6

Instruction Following 1

Long Context 1

BenchmarkGPT-5.5MARGINGPT-6 Luna
AA-LCR84.383.3

Math 1

BenchmarkGPT-5.5MARGINGPT-6 Luna
LiveBench · Mathematics95.9 ◆◆ 89.1

Multimodal 1

BenchmarkGPT-5.5MARGINGPT-6 Luna
MMMU-Pro79.975.5

Other shared benchmarks 7

BenchmarkGPT-5.5MARGINGPT-6 Luna
ARC-AGI-195 ◆◆ 86.7
livebench_data_analysis81.6 ◆◆ 73.4
livebench_language87.4 ◆◆ 73.8

Each score is the one the model's own page shows — the most authoritative tracked result for that benchmark, on the benchmark's standard methodology (source-attributed). ↓ marks lower-is-better metrics; ◆ an independently verified score. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, direct), otherwise the lowest tracked offer.

Compare GPT-5.5 withALL PAIRINGS →