GPT-5 vs GPT-5.6 Luna
Wins 1 of 7 areas
Facts
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Math · Long documents
GPT-5.6 Luna is the stronger all-rounder.GPT-5 is better at facts.
Scores updated · 37 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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.
- CodingWriting and fixing software05GPT-5.6 Luna5 of 5 tests
- ReasoningHard problems that need careful thinking14GPT-5.6 Luna4 of 5 tests
- AgentsCarrying out multi-step tasks on its own03GPT-5.6 Luna3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video03GPT-5.6 Luna3 of 3 tests
- MathCompetition and research-level math02GPT-5.6 Luna2 of 2 tests
- Long documentsFinding answers in very long texts01GPT-5.6 Luna1 of 1 test
- FactsGetting facts right instead of making them up21GPT-52 of 3 tests
Long documents rests on a single test.
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-5.6 Luna costs 88% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GPT-5.6 Luna pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+49.7points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+45.7points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+36.5points ahead
Where GPT-5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+10.4points ahead
- Common-sense trick questionsSimpleBench+9.9points ahead
- Short factual questions, answered correctlySimpleQA Verified+9.1points ahead
Every test, side by side
All 37 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.6 Luna
- Terminal-Bench 2.1GPT-5.6 Luna by 45.735.280.9+45.7
- SWE-bench ProGPT-5.6 Luna by 20.941.862.7+20.9
- SWE-bench VerifiedGPT-5.6 Luna by 18.174.993+18.1
- SciCodeGPT-5.6 Luna by 10.742.953.6+10.7
- LMArena · WebDevGPT-5.6 Luna by 102 rating points14171519+102 rating
AgentsGPT-5.6 Luna
- BrowseCompGPT-5.6 Luna by 28.454.983.3+28.4
- GDPValGPT-5.6 Luna by 27.120.948+27.1
- τ-Bench V3 · BankingGPT-5.6 Luna by 922.131.1+9
ReasoningGPT-5.6 Luna
- ARC-AGI-2GPT-5.6 Luna by 49.79.959.6+49.7
- CritPtGPT-5.6 Luna by 14.95.720.6+14.9
- Humanity's Last ExamGPT-5.6 Luna by 1128.539.5+11
- SimpleBenchGPT-5 by 9.956.746.8+9.9
- GPQA DiamondGPT-5.6 Luna by 5.785.491.1+5.7
FactsGPT-5
- AA-Omniscience · Non-hallucinationGPT-5 by 10.417.87.4+10.4
- SimpleQA VerifiedGPT-5 by 9.150.141+9.1
- AA-Omniscience · AccuracyGPT-5.6 Luna by 2.440.342.7+2.4
Images and chartsGPT-5.6 Luna
- LMArena · VisionGPT-5.6 Luna by 48 rating points12081256+48 rating
- MMMU-ProGPT-5.6 Luna by 4.474.278.6+4.4
- CharXiv (RQ)GPT-5.6 Luna by 1.681.182.7+1.6
MathGPT-5.6 Luna
- FrontierMath Tier 4GPT-5.6 Luna by 36.521.958.5+36.5
- FrontierMath Tiers 1-3 (v2)GPT-5.6 Luna by 26.755.482.1+26.7
Other results15 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.
- FrontierMath (overall)GPT-5.6 Luna by 52.326.378.6+52.3
- Artificial Analysis Coding IndexGPT-5.6 Luna by 33.637.871.4+33.6
- HealthBench HardGPT-5.6 Luna by 30.41.632+30.4
- ARC-AGI-1GPT-5.6 Luna by 2565.790.7+25
- AA Agentic IndexGPT-5.6 Luna by 16.226.542.7+16.2
- AA IntelligenceGPT-5.6 Luna by 14.32337.3+14.3
- HLE (with tools)GPT-5.6 Luna by 13.735.248.9+13.7
- HealthBenchGPT-5 by 11.467.255.8+11.4
- HealthBench Professional length-adjustedGPT-5.6 Luna by 9.546.255.7+9.5
- ERQAGPT-5 by 9.265.756.5+9.2
- VideoMMMUGPT-5 by 3.584.681.1+3.5
- HealthBench Hard length-adjustedGPT-5 by 2.734.732+2.7
- HealthBench length-adjustedGPT-5 by 1.957.755.8+1.9
- AA-OmniscienceGPT-5 by 1.6-8.7-10.3+1.6
- HealthBench Consensus length-adjustedtie95.695.1tie
Questions people ask
Which is better, GPT-5 or GPT-5.6 Luna?
GPT-5.6 Luna wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, math and long documents. GPT-5 wins facts.
Which is better for coding?
GPT-5.6 Luna. It wins 5 of the 5 coding tests both models report; GPT-5 wins none.
Which is cheaper?
GPT-5 costs $1.25 per million input tokens and $10.00 per million output tokens; GPT-5.6 Luna costs $0.20 and $1.20. That makes GPT-5.6 Luna about 88% cheaper for the same work.
How do you compare the two?
We use the 37 benchmark tests both models have published scores on. The verdict counts the 22 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 15 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.