GPT-5.6 Luna vs Kimi K2.5
Wins 6 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Long documents
Wins 1 of 7 areas
Images and charts
GPT-5.6 Luna is the stronger all-rounder.Kimi K2.5 is better at images and charts.
Scores updated · 38 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 software50GPT-5.6 Luna5 of 5 tests
- AgentsCarrying out multi-step tasks on its own50GPT-5.6 Luna5 of 5 tests
- ReasoningHard problems that need careful thinking40GPT-5.6 Luna4 of 4 tests
- MathCompetition and research-level math20GPT-5.6 Luna2 of 2 tests
- FactsGetting facts right instead of making them up21GPT-5.6 Luna2 of 3 tests
- Long documentsFinding answers in very long texts10GPT-5.6 Luna1 of 1 test
- Images and chartsUnderstanding pictures, charts and video23Kimi K2.53 of 5 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 61% 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+47.8points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+35.2points ahead
- Real work tasks from 44 professionsGDPVal+31points ahead
Where Kimi K2.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+26.9points ahead
- Math problems shown in pictures and chartsMathVista+13.2points ahead
- Questions about short and long videosVideo-MME+5.8points ahead
Every test, side by side
All 38 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 35.280.945.7+35.2
- SWE-bench VerifiedGPT-5.6 Luna by 16.29376.8+16.2
- SWE-bench ProGPT-5.6 Luna by 1262.750.7+12
- LMArena · WebDevGPT-5.6 Luna by 83 rating points15191436+83 rating
- SciCodeGPT-5.6 Luna by 4.653.649+4.6
AgentsGPT-5.6 Luna
- GDPValGPT-5.6 Luna by 314817+31
- AA ApexAgentsGPT-5.6 Luna by 24.335.811.5+24.3
- τ-Bench V3 · BankingGPT-5.6 Luna by 16.931.114.2+16.9
- OSWorld-VerifiedGPT-5.6 Luna by 9.372.663.3+9.3
- BrowseCompGPT-5.6 Luna by 8.483.374.9+8.4
ReasoningGPT-5.6 Luna
- ARC-AGI-2GPT-5.6 Luna by 47.859.611.8+47.8
- CritPtGPT-5.6 Luna by 17.520.63.1+17.5
- Humanity's Last ExamGPT-5.6 Luna by 8.839.530.7+8.8
- GPQA DiamondGPT-5.6 Luna by 3.291.187.9+3.2
FactsGPT-5.6 Luna
- AA-Omniscience · Non-hallucinationKimi K2.5 by 26.97.434.3+26.9
- AA-Omniscience · AccuracyGPT-5.6 Luna by 7.542.735.2+7.5
- SimpleQA VerifiedGPT-5.6 Luna by 6.74134.3+6.7
Images and chartsKimi K2.5
- MathVistaKimi K2.5 by 13.276.990.1+13.2
- Video-MMEKimi K2.5 by 5.881.687.4+5.8
- CharXiv (RQ)GPT-5.6 Luna by 5.282.777.5+5.2
- MMMU-ProGPT-5.6 Luna by 3.278.675.4+3.2
- LMArena · VisionKimi K2.5 by 13 rating points12561269+13 rating
MathGPT-5.6 Luna
- HMMT Feb. 2026GPT-5.6 Luna by 11.498.587.1+11.4
- AIME 2026GPT-5.6 Luna by 1.897.695.8+1.8
Other results13 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.
- GDPval-AA v2GPT-5.6 Luna by 521 rating points15301009+521 rating
- ToolathlonGPT-5.6 Luna by 25.653.427.8+25.6
- ARC-AGI-1GPT-5.6 Luna by 25.490.765.3+25.4
- Artificial Analysis Coding IndexGPT-5.6 Luna by 24.671.446.8+24.6
- AA Agentic IndexGPT-5.6 Luna by 2142.721.7+21
- SimpleVQAKimi K2.5 by 1853.271.2+18
- LVBenchKimi K2.5 by 1758.975.9+17
- AA IntelligenceGPT-5.6 Luna by 13.837.323.5+13.8
- SWEBench Pro PublicGPT-5.6 Luna by 1262.750.7+12
- τ³-Bench BankingGPT-5.6 Luna by 10.124.314.2+10.1
- VideoMMMUKimi K2.5 by 5.581.186.6+5.5
- AA-OmniscienceKimi K2.5 by 3-10.3-7.3+3
- HLE (with tools)Kimi K2.5 by 1.348.950.2+1.3
Questions people ask
Which is better, GPT-5.6 Luna or Kimi K2.5?
GPT-5.6 Luna wins six of the seven areas where both have results: coding, agents, reasoning, facts, math and long documents. Kimi K2.5 wins images and charts.
Which is better for coding?
GPT-5.6 Luna. It wins 5 of the 5 coding tests both models report; Kimi K2.5 wins none.
Which is cheaper?
GPT-5.6 Luna costs $0.20 per million input tokens and $1.20 per million output tokens; Kimi K2.5 costs $0.60 and $3.00. That makes GPT-5.6 Luna about 61% cheaper for the same work.
How do you compare the two?
We use the 38 benchmark tests both models have published scores on. The verdict counts the 25 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 13 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.