Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Claude Opus 4.6 vs GPT-5.6 Luna

40 SHARED BENCHMARKS

Across 40 shared benchmarks, Claude Opus 4.6 scores higher on 23 and GPT-5.6 Luna on 17. The widest gap is AA-Omniscience · Non-hallucination, where Claude Opus 4.6 scores 37.2 against 7.4. GPT-5.6 Luna is the cheaper of the two on tracked API pricing ($0.20 against $5.00 per million input tokens).

ANTHROPICVSOPENAI40 SHARED23–17 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerAnthropicOpenAI
Released
Price per 1M tokens input / output$5.00 / $25.00$0.20 / $1.20
Cost of 1M in + 1M out$30.00$1.40 21.4× less
Head-to-head of 40 shared benchmarks23 wins17 wins
Scores tracked independently verified231 55 ◆201 43 ◆

Release dates per Artificial Analysis. Prices: Anthropic's own price page for Claude Opus 4.6; Artificial Analysis for GPT-5.6 Luna. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaClaude Opus 4.6WINSGPT-5.6 Luna
Reasoning52Claude Opus 4.6 leads 5 of 7 · widest: SimpleBench 67.6 vs 46.8
Coding24GPT-5.6 Luna leads 4 of 6 · widest: SWE-bench Pro 53.4 vs 62.7
Agentic31Claude Opus 4.6 leads 3 of 4 · widest: GDPVal 55.9 vs 47.2
Factuality30Claude Opus 4.6 leads 3 of 3 · widest: AA-Omniscience · Non-hallucination 37.2 vs 7.4
Instruction Following10Claude Opus 4.6 leads 1 of 1 · widest: LiveBench · Instruction Following 63.3 vs 60.1
Long Context11Even, 1–1 of 2
Math14GPT-5.6 Luna leads 4 of 5 · widest: FrontierMath Tier 4 26.8 vs 58.5
Multimodal12GPT-5.6 Luna leads 2 of 3 · widest: CharXiv (RQ) 77.4 vs 82.7

Biggest gaps

Claude Opus 4.6 pulls furthest ahead on

  1. AA-Omniscience · Non-hallucination37.2 vs 7.4
  2. SimpleBench67.6 vs 46.8
  3. GDPVal55.9 vs 47.2

GPT-5.6 Luna pulls furthest ahead on

  1. FrontierMath Tier 458.5 vs 26.8
  2. CritPt20.6 vs 12.6
  3. FrontierMath Tiers 1-3 (v2)82.1 vs 66

Every shared benchmark40 · grouped by area

Reasoning 7

ARC-AGI-30.5 ◆0.2
ARC-AGI-268.8◆ 59.6
CritPt12.620.6
GPQA Diamond89.691.1
LiveBench · Reasoning88.7 ◆◆ 85.6
SimpleBench67.6 ◆◆ 46.8

Coding 6

Agentic 4

Instruction Following 1

Math 5

AIME 202696.7 ◆97.6
HMMT Feb. 202696.2 ◆98.5
LiveBench · Mathematics89.3 ◆◆ 87.2

Multimodal 3

LMArena · Vision1316 ◆◆ 1259
CharXiv (RQ)77.482.7
MMMU-Pro75.478.6

Other shared benchmarks 9

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 (anthropic-official, direct), otherwise the lowest tracked offer.

Compare Claude Opus 4.6 withALL PAIRINGS →

Compare GPT-5.6 Luna withALL PAIRINGS →