Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Claude Opus 4.7 vs GPT-5.6 Luna

44 SHARED BENCHMARKS

Across 44 shared benchmarks, Claude Opus 4.7 scores higher on 29 and GPT-5.6 Luna on 14, with 1 level. The widest gap is AA-Omniscience · Non-hallucination, where Claude Opus 4.7 scores 57.7 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).

ANTHROPICVSOPENAI44 SHARED29–14 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 44 shared benchmarks29 wins14 wins
Scores tracked independently verified167 39 ◆201 43 ◆

Release dates per Artificial Analysis. Prices: Artificial Analysis for Claude Opus 4.7; OpenAI's own price page for GPT-5.6 Luna. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaClaude Opus 4.7WINSGPT-5.6 Luna
Reasoning51Claude Opus 4.7 leads 5 of 7 · widest: SimpleBench 61.7 vs 46.8
Coding52Claude Opus 4.7 leads 5 of 7 · widest: LMArena · WebDev 1557 vs 1520
Agentic33Even, 3–3 of 6
Factuality30Claude Opus 4.7 leads 3 of 3 · widest: AA-Omniscience · Non-hallucination 57.7 vs 7.4
Instruction Following10Claude Opus 4.7 leads 1 of 1 · widest: LiveBench · Instruction Following 66.7 vs 60.1
Long Context02GPT-5.6 Luna leads 2 of 2 · widest: MRCR v2 8 needle 128k (average) 59.3 vs 74.8
Math14GPT-5.6 Luna leads 4 of 5 · widest: FrontierMath Tier 4 31.7 vs 58.5
Multimodal30Claude Opus 4.7 leads 3 of 3 · widest: CharXiv (RQ) 91 vs 82.7

Biggest gaps

Claude Opus 4.7 pulls furthest ahead on

  1. AA-Omniscience · Non-hallucination57.7 vs 7.4
  2. SimpleBench61.7 vs 46.8
  3. ARC-AGI-275.8 vs 59.6

GPT-5.6 Luna pulls furthest ahead on

  1. FrontierMath Tier 458.5 vs 31.7
  2. CritPt20.6 vs 12
  3. MRCR v2 8 needle 128k (average)74.8 vs 59.3

Every shared benchmark44 · grouped by area

Reasoning 7

ARC-AGI-30.2 ◆0.2
ARC-AGI-275.8◆ 59.6
CritPt1220.6
GPQA Diamond91.491.1
LiveBench · Reasoning87.2 ◆◆ 85.6
SimpleBench61.7 ◆◆ 46.8

Coding 7

Instruction Following 1

Long Context 2

Math 5

AIME 202695.8 ◆97.6
HMMT Feb. 202693.9 ◆98.5
LiveBench · Mathematics92.8 ◆◆ 87.2

Multimodal 3

LMArena · Vision1314 ◆◆ 1259
MMMU-Pro78.878.6

Other shared benchmarks 10

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, openai-official), otherwise the lowest tracked offer.

Compare Claude Opus 4.7 withALL PAIRINGS →

Compare GPT-5.6 Luna withALL PAIRINGS →