GPT-5.6 Luna vs Inkling-Small
Wins 7 of 7 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents
Wins 0 of 7 areas
—
GPT-5.6 Luna is the stronger all-rounder.
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 software50GPT-5.6 Luna5 of 5 tests
- AgentsCarrying out multi-step tasks on its own40GPT-5.6 Luna4 of 4 tests
- ReasoningHard problems that need careful thinking40GPT-5.6 Luna4 of 4 tests
- MathCompetition and research-level math40GPT-5.6 Luna4 of 4 tests
- Images and chartsUnderstanding pictures, charts and video30GPT-5.6 Luna3 of 3 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
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 7% 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-5.6 Luna pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+41.4points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+35.8points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+25.8points ahead
Where Inkling-Small pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+29.6points 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 25.880.955.1+25.8
- SWE-bench VerifiedGPT-5.6 Luna by 12.89380.2+12.8
- LMArena · WebDevGPT-5.6 Luna by 111 rating points15201409+111 rating
- SWE-bench ProGPT-5.6 Luna by 6.862.755.9+6.8
- SciCodeGPT-5.6 Luna by 3.953.649.7+3.9
AgentsGPT-5.6 Luna
- GDPValGPT-5.6 Luna by 1748.231.2+17
- τ-Bench V3 · BankingGPT-5.6 Luna by 12.331.118.8+12.3
- Terminal-Bench 4.0GPT-5.6 Luna by 10.611.61+10.6
- BrowseCompGPT-5.6 Luna by 5.983.377.4+5.9
ReasoningGPT-5.6 Luna
- ARC-AGI-2GPT-5.6 Luna by 19.559.640.1+19.5
- CritPtGPT-5.6 Luna by 12.320.68.3+12.3
- Humanity's Last ExamGPT-5.6 Luna by 6.239.533.3+6.2
- GPQA DiamondGPT-5.6 Luna by 1.691.189.5+1.6
FactsGPT-5.6 Luna
- AA-Omniscience · Non-hallucinationInkling-Small by 29.67.437+29.6
- SimpleQA VerifiedGPT-5.6 Luna by 20.44120.6+20.4
- AA-Omniscience · AccuracyGPT-5.6 Luna by 9.542.733.2+9.5
Images and chartsGPT-5.6 Luna
- CharXiv (RQ)GPT-5.6 Luna by 5.382.777.4+5.3
- MMMU-ProGPT-5.6 Luna by 4.678.674+4.6
- LMArena · VisionGPT-5.6 Luna by 26 rating points12601234+26 rating
MathGPT-5.6 Luna
- FrontierMath Tier 4GPT-5.6 Luna by 41.458.517.1+41.4
- FrontierMath Tiers 1-3 (v2)GPT-5.6 Luna by 35.882.146.3+35.8
- HMMT Feb. 2026GPT-5.6 Luna by 8.398.590.2+8.3
- AIME 2026GPT-5.6 Luna by 2.197.695.5+2.1
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 292 rating points15301238+292 rating
- Artificial Analysis Coding IndexGPT-5.6 Luna by 18.571.452.9+18.5
- AA Agentic IndexGPT-5.6 Luna by 17.742.725+17.7
- Toolathlon VerifiedGPT-5.6 Luna by 13.567.954.4+13.5
- AA IntelligenceGPT-5.6 Luna by 11.637.325.7+11.6
- Artificial AnalysisGPT-5.6 Luna by 115140+11
- τ³-Bench BankingGPT-5.6 Luna by 8.824.315.5+8.8
- SWEBench Pro PublicGPT-5.6 Luna by 6.862.755.9+6.8
- ARC-AGI-1GPT-5.6 Luna by 6.790.784+6.7
- BrowseComp (context management)GPT-5.6 Luna by 6.68477.4+6.6
- AA-OmniscienceInkling-Small by 1.4-10.3-8.9+1.4
- HLE (with tools)GPT-5.6 Luna by 1.148.947.8+1.1
- ToolathlonInkling-Small by 153.454.4+1
Questions people ask
Which is better, GPT-5.6 Luna or Inkling-Small?
GPT-5.6 Luna wins all seven areas where both have results: coding, agents, reasoning, facts, images and charts, math and long documents. Inkling-Small wins none.
Which is better for coding?
GPT-5.6 Luna. It wins 5 of the 5 coding tests both models report; Inkling-Small wins none.
Which is cheaper?
GPT-5.6 Luna costs $0.20 per million input tokens and $1.20 per million output tokens; Inkling-Small costs $0.30 and $1.20. That makes GPT-5.6 Luna about 7% 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 24 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.