Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Claude Fable 5 vs GPT-5.6 Luna

50 SHARED BENCHMARKS

Across 50 shared benchmarks, Claude Fable 5 scores higher on 48 and GPT-5.6 Luna on 2. The widest gap is AA-Omniscience · Non-hallucination, where Claude Fable 5 scores 36.4 against 7.4. GPT-5.6 Luna is the cheaper of the two on tracked API pricing ($0.20 against $10.00 per million input tokens).

ANTHROPICVSOPENAI50 SHARED48–2 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerAnthropicOpenAI
Released
Price per 1M tokens input / output$10.00 / $50.00$0.20 / $1.20
Cost of 1M in + 1M out$60.00$1.40 42.9× less
Head-to-head of 50 shared benchmarks48 wins2 wins
Scores tracked independently verified154 31 ◆183 43 ◆

Release dates per Artificial Analysis. Prices: Artificial Analysis for Claude Fable 5; Artificial Analysis for GPT-5.6 Luna. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaClaude Fable 5WINSGPT-5.6 Luna
Reasoning60Claude Fable 5 leads 6 of 6 · widest: SimpleBench 81.9 vs 46.8
Coding70Claude Fable 5 leads 7 of 7 · widest: LiveBench · Agentic Coding 62.2 vs 48.4
Agentic60Claude Fable 5 leads 6 of 6 · widest: Terminal-Bench 4.0 42.4 vs 11.6
Factuality30Claude Fable 5 leads 3 of 3 · widest: AA-Omniscience · Non-hallucination 36.4 vs 7.4
Instruction Following10Claude Fable 5 leads 1 of 1 · widest: LiveBench · Instruction Following 75.8 vs 60.1
Long Context01GPT-5.6 Luna leads 1 of 1
Math30Claude Fable 5 leads 3 of 3 · widest: FrontierMath Tier 4 90.2 vs 58.5
Multimodal20Claude Fable 5 leads 2 of 2 · widest: MMMU-Pro 84.2 vs 78.6

Biggest gaps

Claude Fable 5 pulls furthest ahead on

  1. AA-Omniscience · Non-hallucination36.4 vs 7.4
  2. Terminal-Bench 4.042.4 vs 11.6
  3. SimpleBench81.9 vs 46.8

GPT-5.6 Luna pulls furthest ahead on

No ratified-area lead of 3 points or more.

Every shared benchmark50 · grouped by area

Reasoning 6

ARC-AGI-289.2◆ 59.6
CritPt28.620.6
GPQA Diamond92.691.1
LiveBench · Reasoning89.7 ◆◆ 85.6
SimpleBench81.9 ◆◆ 46.8

Agentic 6

BrowseComp87.483.3
GDPVal54.847.2
Terminal-Bench 4.042.411.6

Instruction Following 1

Long Context 1

AA-LCR82.383.7

Multimodal 2

LMArena · Vision1326 ◆◆ 1259
MMMU-Pro84.278.6

Other shared benchmarks 21

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 Claude Fable 5 withALL PAIRINGS →