Claude Fable 5.2 vs GPT-6 Luna
Wins 5 of 6 areas
Coding · Agents · Reasoning · Facts · Math
Wins 0 of 6 areas
—
Claude Fable 5.2 is the stronger all-rounder.GPT-6 Luna is cheaper.
Scores updated · 22 tests both models report · How we compare
Where each one wins
Tests won in each of the six areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking40Claude Fable 5.24 of 4 tests
- MathCompetition and research-level math20Claude Fable 5.22 of 2 tests
- CodingWriting and fixing software10Claude Fable 5.21 of 1 test
- AgentsCarrying out multi-step tasks on its own10Claude Fable 5.21 of 1 test
- FactsGetting facts right instead of making them up10Claude Fable 5.21 of 1 test
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
Coding, agents, facts and long documents rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
GPT-6 Luna costs 99% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Claude Fable 5.2 pulls ahead
- Solves new puzzle games it has never seenARC-AGI-3+79.2points ahead
- Short factual questions, answered correctlySimpleQA Verified+52.6points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+47points ahead
Where GPT-6 Luna pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 22 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningClaude Fable 5.2
- ARC-AGI-3Claude Fable 5.2 by 79.279.30.1+79.2
- Humanity's Last ExamClaude Fable 5.2 by 35.874.338.5+35.8
- ARC-AGI-2Claude Fable 5.2 by 33.292.559.3+33.2
- CritPtClaude Fable 5.2 by 1635.419.4+16
MathClaude Fable 5.2
- FrontierMath Tier 4Claude Fable 5.2 by 43.910056.1+43.9
- FrontierMath Tiers 1-3 (v2)Claude Fable 5.2 by 16.195.179+16.1
Other results12 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- Terminal-Bench-Science 0.1Claude Fable 5.2 by 64.773.38.6+64.7
- HLE-DiamondClaude Fable 5.2 by 5880.222.2+58
- Terminal-Bench 4.0Claude Fable 5.2 by 51.267.616.4+51.2
- GDPval-AA 2.1Claude Fable 5.2 by 394 rating points18311437+394 rating
- CritPt v2 HardClaude Fable 5.2 by 3476.542.5+34
- ARC-AGI-1Claude Fable 5.2 by 10.897.586.7+10.8
- DeepSWE 1.1Claude Fable 5.2 by 7.774.366.6+7.7
- HealthBench Professional length-adjustedClaude Fable 5.2 by 6.967.760.8+6.9
- HealthBench ProfessionalClaude Fable 5.2 by 5.867.761.9+5.8
- HealthBench length-adjustedClaude Fable 5.2 by 5.159.654.5+5.1
- HealthBenchClaude Fable 5.2 by 559.654.6+5
- GDP (Surge AI)Claude Fable 5.2 by 1.424.222.8+1.4
Questions people ask
Which is better, Claude Fable 5.2 or GPT-6 Luna?
Claude Fable 5.2 wins five of the six areas where both have results: coding, agents, reasoning, facts and math. GPT-6 Luna wins none, but costs 99% less. They are level on long documents.
Which is better for coding?
Claude Fable 5.2. It wins the one coding test both models report.
Which is cheaper?
Claude Fable 5.2 costs $8.00 per million input tokens and $40.00 per million output tokens; GPT-6 Luna costs $0.10 and $0.50. That makes GPT-6 Luna about 99% cheaper for the same work.
How do you compare the two?
We use the 22 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 12 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.