Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

GPT-5.2 vs GPT-6 Luna

24 SHARED BENCHMARKS

Across 24 shared benchmarks, GPT-5.2 scores higher on 9 and GPT-6 Luna on 15. The widest gap is CritPt, where GPT-6 Luna scores 19.4 against 11.6. GPT-6 Luna is the cheaper of the two on tracked API pricing ($0.10 against $1.75 per million input tokens).

OPENAIVSOPENAI24 SHARED9–15 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerOpenAIOpenAI
Released
Price per 1M tokens input / output$1.75 / $14.00$0.10 / $0.50
Cost of 1M in + 1M out$15.75$0.60 26.3× less
Head-to-head of 24 shared benchmarks9 wins15 wins
Scores tracked independently verified204 64 ◆121 33 ◆

Release dates per Artificial Analysis. Prices: Artificial Analysis for GPT-5.2; Artificial Analysis for GPT-6 Luna. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaGPT-5.2WINSGPT-6 Luna
Reasoning13GPT-6 Luna leads 3 of 4 · widest: CritPt 11.6 vs 19.4
Coding04GPT-6 Luna leads 4 of 4 · widest: LMArena · WebDev 1415 vs 1593
Agentic10GPT-5.2 leads 1 of 1 · widest: GDPVal 48.3 vs 43.4
Factuality11Even, 1–1 of 2
Instruction Following10GPT-5.2 leads 1 of 1 · widest: LiveBench · Instruction Following 61.8 vs 55.9
Long Context01GPT-6 Luna leads 1 of 1
Math10GPT-5.2 leads 1 of 1 · widest: LiveBench · Mathematics 93.2 vs 89.1
Multimodal01GPT-6 Luna leads 1 of 1

Biggest gaps

GPT-5.2 pulls furthest ahead on

  1. GDPVal48.3 vs 43.4
  2. LiveBench · Instruction Following61.8 vs 55.9
  3. LiveBench · Mathematics93.2 vs 89.1

GPT-6 Luna pulls furthest ahead on

  1. CritPt19.4 vs 11.6
  2. AA-Omniscience · Non-hallucination23.3 vs 18.8
  3. LMArena · WebDev1593 vs 1415

Every shared benchmark24 · grouped by area

Reasoning 4

BenchmarkGPT-5.2MARGINGPT-6 Luna
ARC-AGI-252.9◆ 59.3
CritPt11.619.4
LiveBench · Reasoning83.2 ◆◆ 81.8

Coding 4

BenchmarkGPT-5.2MARGINGPT-6 Luna
LMArena · WebDev1415 ◆◆ 1593
LiveBench · Coding76.1 ◆◆ 79
SciCode52.154.6

Agentic 1

BenchmarkGPT-5.2MARGINGPT-6 Luna
GDPVal48.343.4

Instruction Following 1

Long Context 1

BenchmarkGPT-5.2MARGINGPT-6 Luna
AA-LCR82.783.3

Math 1

BenchmarkGPT-5.2MARGINGPT-6 Luna
LiveBench · Mathematics93.2 ◆◆ 89.1

Multimodal 1

BenchmarkGPT-5.2MARGINGPT-6 Luna
MMMU-Pro74.675.5

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-5.2 withALL PAIRINGS →