GLM-5.3 vs GPT-5.6 Luna
Wins 4 of 7 areas
Coding · Agents · Reasoning · Following instructions
Wins 2 of 7 areas
Math · Long documents
GLM-5.3 wins more areas, narrowly.GPT-5.6 Luna is cheaper and better at math.
Scores updated · 39 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.
- AgentsCarrying out multi-step tasks on its own40GLM-5.34 of 4 tests
- CodingWriting and fixing software41GLM-5.34 of 5 tests
- ReasoningHard problems that need careful thinking21GLM-5.32 of 5 tests · 2 ties
- Following instructionsDoing exactly what it is asked10GLM-5.31 of 1 test
- MathCompetition and research-level math02GPT-5.6 Luna2 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts01GPT-5.6 Luna1 of 1 test
- FactsGetting facts right instead of making them up11Even1 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-5.6 Luna costs 76% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GLM-5.3 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+63points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+30.3points ahead
- Common-sense trick questionsSimpleBench+19.4points ahead
Where GPT-5.6 Luna pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+29.2points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+13.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+8.8points ahead
Every test, side by side
All 39 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGLM-5.3
- LiveBench · Agentic CodingGLM-5.3 by 12.560.948.4+12.5
- LMArena · WebDevGLM-5.3 by 103 rating points16221519+103 rating
- SciCodeGLM-5.3 by 5.45953.6+5.4
- LiveBench · CodingGPT-5.6 Luna by 3.97982.9+3.9
- Terminal-Bench 2.1GLM-5.3 by 383.980.9+3
AgentsGLM-5.3
- Terminal-Bench 4.0GLM-5.3 by 30.341.911.6+30.3
- τ-Bench V3 · BankingGLM-5.3 by 19.250.331.1+19.2
- GDPValGLM-5.3 by 9.557.548+9.5
- AA IT-Bench SREGLM-5.3 by 5.846.140.3+5.8
ReasoningGLM-5.3
- SimpleBenchGLM-5.3 by 19.466.246.8+19.4
- Humanity's Last ExamGLM-5.3 by 2.842.339.5+2.8
- CritPtGPT-5.6 Luna by 1.519.120.6+1.5
- GPQA Diamondtie91.791.1tie
- LiveBench · Reasoningtie85.885.6tie
FactsEven
- AA-Omniscience · Non-hallucinationGLM-5.3 by 6370.47.4+63
- AA-Omniscience · AccuracyGPT-5.6 Luna by 8.833.942.7+8.8
- SimpleQA Verifiedtie4141tie
MathGPT-5.6 Luna
- FrontierMath Tier 4GPT-5.6 Luna by 29.229.358.5+29.2
- FrontierMath Tiers 1-3 (v2)GPT-5.6 Luna by 13.368.882.1+13.3
- LiveBench · Mathematicstie87.987.2tie
Following instructionsGLM-5.3
- LiveBench · Instruction FollowingGLM-5.3 by 9.269.360.1+9.2
Other results17 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.
- AA-OmniscienceGLM-5.3 by 24.614.3-10.3+24.6
- Terminal-Bench 4.0GLM-5.3 by 24.541.817.3+24.5
- GDPval-AA v2GLM-5.3 by 239 rating points17691530+239 rating
- Agents' Last ExamGPT-5.6 Luna by 21.828.550.3+21.8
- ToolathlonGLM-5.3 by 19.67353.4+19.6
- HLE (with tools)GLM-5.3 by 13.662.548.9+13.6
- GDP (Surge AI)GPT-5.6 Luna by 11.511.222.7+11.5
- AA Agentic IndexGLM-5.3 by 10.753.442.7+10.7
- livebench_data_analysisGPT-5.6 Luna by 7.870.278+7.8
- AA IntelligenceGLM-5.3 by 7.544.837.3+7.5
- livebench_languageGLM-5.3 by 7.379.972.6+7.3
- Toolathlon VerifiedGLM-5.3 by 5.17367.9+5.1
- Artificial Analysis Coding IndexGLM-5.3 by 3.474.871.4+3.4
- HealthBench ProfessionalGPT-5.6 Luna by 2.952.855.7+2.9
- HealthBenchGPT-5.6 Luna by 1.354.555.8+1.3
- DeepSWEtie66.967.2tie
- DeepSWE 1.1tie66.967tie
Questions people ask
Which is better, GLM-5.3 or GPT-5.6 Luna?
GLM-5.3 wins four of the seven areas where both have results: coding, agents, reasoning and following instructions. GPT-5.6 Luna wins math and long documents, and costs 76% less. They are level on facts.
Which is better for coding?
GLM-5.3. It wins 4 of the 5 coding tests both models report; GPT-5.6 Luna wins 1.
Which is cheaper?
GLM-5.3 costs $1.40 per million input tokens and $4.40 per million output tokens; GPT-5.6 Luna costs $0.20 and $1.20. That makes GPT-5.6 Luna about 76% cheaper for the same work.
How do you compare the two?
We use the 39 benchmark tests both models have published scores on. The verdict counts the 22 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 17 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.