GPT-5.6 Sol vs GPT-6.1 Sol
Wins 2 of 8 areas
Coding · Long documents
Wins 6 of 8 areas
Agents · Reasoning · Facts · Images and charts · Math · Following instructions
GPT-6.1 Sol is the stronger all-rounder.GPT-5.6 Sol is better at coding.
Both rank among the ten best models we track in reasoning and math.
Scores updated · 41 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 thinking04GPT-6.1 Sol4 of 6 tests · 2 ties
- FactsGetting facts right instead of making them up03GPT-6.1 Sol3 of 3 tests
- AgentsCarrying out multi-step tasks on its own02GPT-6.1 Sol2 of 3 tests · 1 tie
- MathCompetition and research-level math02GPT-6.1 Sol2 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video01GPT-6.1 Sol1 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked01GPT-6.1 Sol1 of 1 test
- CodingWriting and fixing software31GPT-5.6 Sol3 of 4 tests
- Long documentsFinding answers in very long texts10GPT-5.6 Sol1 of 1 test
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-6.1 Sol costs 50% 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-6.1 Sol pulls ahead
- Solves new puzzle games it has never seenARC-AGI-3+44.9points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+37.9points ahead
- Long professional tasks in banking, consulting and lawAA ApexAgents+20.1points ahead
Where GPT-5.6 Sol pulls ahead
- Programming problems, refreshed regularlyLiveBench · Coding+3.5points ahead
- Code for real scientific research problemsSciCode+2.9points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+1.6points ahead
Every test, side by side
All 41 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.6 Sol
- LMArena · WebDevGPT-6.1 Sol by 128 rating points16151743+128 rating
- LiveBench · CodingGPT-5.6 Sol by 3.583.980.4+3.5
- SciCodeGPT-5.6 Sol by 2.957.154.2+2.9
- LiveBench · Agentic CodingGPT-5.6 Sol by 1.656.254.5+1.6
AgentsGPT-6.1 Sol
- AA ApexAgentsGPT-6.1 Sol by 20.139.960+20.1
- Terminal-Bench 4.0GPT-6.1 Sol by 16.239.956.1+16.2
- GDPValtie55.454.7tie
ReasoningGPT-6.1 Sol
- ARC-AGI-3GPT-6.1 Sol by 44.97.852.7+44.9
- SimpleBenchGPT-6.1 Sol by 18.164.882.9+18.1
- Humanity's Last ExamGPT-6.1 Sol by 3.449.552.9+3.4
- ARC-AGI-2GPT-6.1 Sol by 1.792.594.2+1.7
- LiveBench · Reasoningtie91.792.6tie
- CritPttie32.331.7tie
FactsGPT-6.1 Sol
- AA-Omniscience · Non-hallucinationGPT-6.1 Sol by 37.97.845.7+37.9
- SimpleQA VerifiedGPT-6.1 Sol by 4.269.773.9+4.2
- AA-Omniscience · AccuracyGPT-6.1 Sol by 2.759.462.1+2.7
Images and chartsGPT-6.1 Sol
- MMMU-ProGPT-6.1 Sol by 2.683.486+2.6
- LMArena · Visiontie12821286tie
MathGPT-6.1 Sol
- FrontierMath Tier 4GPT-6.1 Sol by 1783100+17
- FrontierMath Tiers 1-3 (v2)GPT-6.1 Sol by 4.689.193.7+4.6
- LiveBench · Mathematicstie96.296.8tie
Following instructionsGPT-6.1 Sol
- LiveBench · Instruction FollowingGPT-6.1 Sol by 2.371.874.2+2.3
Other results18 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 0.1GPT-6.1 Sol by 35.722.458.1+35.7
- HiL-Bench (Tools-allowed)GPT-6.1 Sol by 3032.362.3+30
- Terminal-Bench 4.0GPT-6.1 Sol by 20.937.358.2+20.9
- AA-OmniscienceGPT-6.1 Sol by 19.52241.5+19.5
- OSWorld 2.0GPT-6.1 Sol by 8.862.671.4+8.8
- AA IntelligenceGPT-6.1 Sol by 4.84751.8+4.8
- HealthBench ProfessionalGPT-6.1 Sol by 4.660.565.1+4.6
- HealthBench Professional length-adjustedGPT-6.1 Sol by 3.760.564.2+3.7
- HealthBench HardGPT-6.1 Sol by 3.133.136.2+3.1
- HealthBench Hard length-adjustedGPT-6.1 Sol by 3.133.136.2+3.1
- livebench_data_analysisGPT-6.1 Sol by 2.979.882.7+2.9
- livebench_languageGPT-6.1 Sol by 2.487.790.1+2.4
- DeepSWE 1.1GPT-6.1 Sol by 2.27375.2+2.2
- HealthBenchGPT-6.1 Sol by 1.55758.5+1.5
- HealthBench length-adjustedGPT-6.1 Sol by 1.55758.5+1.5
- HealthBench Consensus length-adjustedtie95.596tie
- GDP (Surge AI)tie30.731tie
- ARC-AGI-1tie96.596.5tie
Questions people ask
Which is better, GPT-5.6 Sol or GPT-6.1 Sol?
GPT-6.1 Sol wins six of the eight areas we test: agents, reasoning, facts, images and charts, math and following instructions. GPT-5.6 Sol wins coding and long documents.
Which is better for coding?
GPT-5.6 Sol. It wins 3 of the 4 coding tests both models report; GPT-6.1 Sol wins 1.
Which is cheaper?
GPT-5.6 Sol costs $4.00 per million input tokens and $20.00 per million output tokens; GPT-6.1 Sol costs $2.00 and $10.00. That makes GPT-6.1 Sol about 50% cheaper for the same work.
How do you compare the two?
We use the 41 benchmark tests both models have published scores on. The verdict counts the 23 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.