GPT-5.6 Luna vs Qwen3.5 122B A10B
Wins 5 of 6 areas
Coding · Agents · Reasoning · Images and charts · Long documents
Wins 0 of 6 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 six 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 thinking30GPT-5.6 Luna3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video32GPT-5.6 Luna3 of 5 tests
- Long documentsFinding answers in very long texts10GPT-5.6 Luna1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
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
- Command-line tasks in a real terminalTerminal-Bench 2.1+33.3points ahead
- Real work tasks from 44 professionsGDPVal+32.3points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+21points ahead
Where Qwen3.5 122B A10B pulls ahead
- Questions that truly need the image to answerMMStar+10.7points ahead
- Math problems shown in pictures and chartsMathVista+10.5points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+5.5points 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 33.380.947.6+33.3
- SWE-bench VerifiedGPT-5.6 Luna by 219372+21
- SWE-bench ProGPT-5.6 Luna by 17.862.744.9+17.8
- LMArena · WebDevGPT-5.6 Luna by 159 rating points15191360+159 rating
- SciCodeGPT-5.6 Luna by 13.953.639.7+13.9
AgentsGPT-5.6 Luna
- GDPValGPT-5.6 Luna by 32.34815.7+32.3
- BrowseCompGPT-5.6 Luna by 19.583.363.8+19.5
- τ-Bench V3 · BankingGPT-5.6 Luna by 15.831.115.3+15.8
- OSWorld-VerifiedGPT-5.6 Luna by 14.672.658+14.6
- Terminal-Bench 4.0GPT-5.6 Luna by 11.611.60+11.6
ReasoningGPT-5.6 Luna
- CritPtGPT-5.6 Luna by 2020.60.6+20
- Humanity's Last ExamGPT-5.6 Luna by 14.339.525.2+14.3
- GPQA DiamondGPT-5.6 Luna by 5.491.185.7+5.4
FactsEven
- AA-Omniscience · AccuracyGPT-5.6 Luna by 18.342.724.4+18.3
- AA-Omniscience · Non-hallucinationQwen3.5 122B A10B by 5.57.412.9+5.5
Images and chartsGPT-5.6 Luna
- MMStarQwen3.5 122B A10B by 10.772.282.9+10.7
- MathVistaQwen3.5 122B A10B by 10.576.987.4+10.5
- CharXiv (RQ)GPT-5.6 Luna by 5.582.777.2+5.5
- MMMU-ProGPT-5.6 Luna by 3.678.675+3.6
- LMArena · VisionGPT-5.6 Luna by 12 rating points12571245+12 rating
Other results16 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 669 rating points1530861+669 rating
- AA Agentic IndexGPT-5.6 Luna by 33.142.79.6+33.1
- AA-OmniscienceGPT-5.6 Luna by 31.2-10.3-41.5+31.2
- Artificial Analysis Coding IndexGPT-5.6 Luna by 25.771.445.7+25.7
- AA IntelligenceGPT-5.6 Luna by 21.737.315.6+21.7
- RefSpatial-BenchQwen3.5 122B A10B by 16.852.569.3+16.8
- LVBenchQwen3.5 122B A10B by 15.558.974.4+15.5
- RealWorldQAQwen3.5 122B A10B by 12.772.485.1+12.7
- τ³-Bench BankingGPT-5.6 Luna by 10.124.314.2+10.1
- HallusionBenchQwen3.5 122B A10B by 8.65967.6+8.6
- SimpleVQAQwen3.5 122B A10B by 8.553.261.7+8.5
- EmbSpatial-BenchQwen3.5 122B A10B by 8.475.583.9+8.4
- MV-BenchQwen3.5 122B A10B by 8.468.276.6+8.4
- ERQAQwen3.5 122B A10B by 5.556.562+5.5
- OCRBench v2_enQwen3.5 122B A10B by 3.564.868.3+3.5
- VideoMMMUtie81.182tie
Questions people ask
Which is better, GPT-5.6 Luna or Qwen3.5 122B A10B?
GPT-5.6 Luna wins five of the six areas where both have results: coding, agents, reasoning, images and charts and long documents. Qwen3.5 122B A10B wins none. They are level on facts.
Which is better for coding?
GPT-5.6 Luna. It wins 5 of the 5 coding tests both models report; Qwen3.5 122B A10B wins none.
Which is cheaper?
GPT-5.6 Luna costs $0.20 per million input tokens and $1.20 per million output tokens; Qwen3.5 122B A10B costs $0.40 and $3.20. That makes GPT-5.6 Luna about 61% 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 21 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 16 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.