GPT-6 Luna vs GPT-6 Sol
Wins 0 of 8 areas
—
Wins 7 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Following instructions
GPT-6 Sol is the stronger all-rounder.GPT-6 Luna is cheaper.
Scores updated · 36 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. 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 thinking05GPT-6 Sol5 of 5 tests
- CodingWriting and fixing software04GPT-6 Sol4 of 4 tests
- FactsGetting facts right instead of making them up03GPT-6 Sol3 of 3 tests
- MathCompetition and research-level math03GPT-6 Sol3 of 3 tests
- AgentsCarrying out multi-step tasks on its own02GPT-6 Sol2 of 2 tests
- Images and chartsUnderstanding pictures, charts and video02GPT-6 Sol2 of 2 tests
- Following instructionsDoing exactly what it is asked01GPT-6 Sol1 of 1 test
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
Long documents and following instructions 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 95% 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 GPT-6 Sol pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+33.9points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+31.3points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+30.3points ahead
Where GPT-6 Luna pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 36 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-6 Sol
- LMArena · WebDevGPT-6 Sol by 96 rating points15771673+96 rating
- SciCodeGPT-6 Sol by 354.657.6+3
- LiveBench · CodingGPT-6 Sol by 2.87981.8+2.8
- LiveBench · Agentic CodingGPT-6 Sol by 1.751.252.9+1.7
AgentsGPT-6 Sol
- Terminal-Bench 4.0GPT-6 Sol by 31.312.643.9+31.3
- GDPValGPT-6 Sol by 3.846.650.4+3.8
ReasoningGPT-6 Sol
- ARC-AGI-2GPT-6 Sol by 30.359.389.6+30.3
- CritPtGPT-6 Sol by 11.519.430.9+11.5
- Humanity's Last ExamGPT-6 Sol by 9.438.547.9+9.4
- LiveBench · ReasoningGPT-6 Sol by 6.981.888.7+6.9
- ARC-AGI-3GPT-6 Sol by 4.50.14.6+4.5
FactsGPT-6 Sol
- SimpleQA VerifiedGPT-6 Sol by 19.341.460.7+19.3
- AA-Omniscience · Non-hallucinationGPT-6 Sol by 16.623.339.9+16.6
- AA-Omniscience · AccuracyGPT-6 Sol by 10.743.854.5+10.7
Images and chartsGPT-6 Sol
- MMMU-ProGPT-6 Sol by 3.279.782.9+3.2
- LMArena · VisionGPT-6 Sol by 22 rating points12111233+22 rating
MathGPT-6 Sol
- FrontierMath Tier 4GPT-6 Sol by 33.956.190+33.9
- FrontierMath Tiers 1-3 (v2)GPT-6 Sol by 10.87989.8+10.8
- LiveBench · MathematicsGPT-6 Sol by 7.389.196.4+7.3
Following instructionsGPT-6 Sol
- LiveBench · Instruction FollowingGPT-6 Sol by 12.755.968.6+12.7
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.
- Terminal-Bench 4.0GPT-6 Sol by 3316.449.4+33
- AA-OmniscienceGPT-6 Sol by 26.40.727.1+26.4
- Terminal-Bench-Science 0.1GPT-6 Sol by 198.627.6+19
- OSWorld 2.0GPT-6 Sol by 11.752.764.4+11.7
- livebench_languageGPT-6 Sol by 11.573.885.3+11.5
- AA IntelligenceGPT-6 Sol by 9.538.147.6+9.5
- ARC-AGI-1GPT-6 Sol by 8.886.795.5+8.8
- livebench_data_analysisGPT-6 Sol by 7.873.481.2+7.8
- Agents' Last ExamGPT-6 Sol by 5.550.956.4+5.5
- GDPval-AA 2.1GPT-6 Sol by 50 rating points14371487+50 rating
- DeepSWE 1.1GPT-6 Sol by 2.266.668.8+2.2
- HealthBench Hard length-adjustedGPT-6 Luna by 1.331.430.1+1.3
- HealthBench length-adjustedGPT-6 Luna by 1.354.553.2+1.3
- HealthBench Consensus length-adjustedtie95.996.2tie
- HealthBench Professional length-adjustedtie60.860.8tie
Questions people ask
Which is better, GPT-6 Luna or GPT-6 Sol?
GPT-6 Sol wins seven of the eight areas we test: coding, agents, reasoning, facts, images and charts, math and following instructions. GPT-6 Luna wins none, but costs 95% less. They are level on long documents.
Which is better for coding?
GPT-6 Sol. It wins 4 of the 4 coding tests both models report; GPT-6 Luna wins none.
Which is cheaper?
GPT-6 Luna costs $0.10 per million input tokens and $0.50 per million output tokens; GPT-6 Sol costs $2.00 and $10.00. That makes GPT-6 Luna about 95% cheaper for the same work.
How do you compare the two?
We use the 36 benchmark tests both models have published scores on. The verdict counts the 21 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.