GLM-5.2 vs GLM-5.3-Flash
Wins 4 of 7 areas
Reasoning · Facts · Math · Following instructions
Wins 3 of 7 areas
Coding · Agents · Long documents
GLM-5.2 wins more areas, narrowly.GLM-5.3-Flash is cheaper and better at agents.
Scores updated · 36 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.
- MathCompetition and research-level math30GLM-5.23 of 3 tests
- ReasoningHard problems that need careful thinking32GLM-5.23 of 5 tests
- FactsGetting facts right instead of making them up21GLM-5.22 of 3 tests
- Following instructionsDoing exactly what it is asked10GLM-5.21 of 1 test
- AgentsCarrying out multi-step tasks on its own04GLM-5.3-Flash4 of 4 tests
- CodingWriting and fixing software02GLM-5.3-Flash2 of 5 tests · 3 ties
- Long documentsFinding answers in very long texts01GLM-5.3-Flash1 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.
GLM-5.3-Flash costs 89% 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-5.2 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+12.2points ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+9.5points ahead
- Fresh competition math problemsLiveBench · Mathematics+8.6points ahead
Every test, side by side
All 36 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGLM-5.3-Flash
- Terminal-Bench 2.1GLM-5.3-Flash by 6.477.984.3+6.4
- LiveBench · Agentic CodingGLM-5.3-Flash by 551.856.8+5
- LiveBench · Codingtie79.779tie
- SciCodetie51.251.6tie
- LMArena · WebDevtie16081609tie
AgentsGLM-5.3-Flash
- Terminal-Bench 4.0GLM-5.3-Flash by 31.8132.8+31.8
- GDPValGLM-5.3-Flash by 13.643.657.2+13.6
- τ-Bench V3 · BankingGLM-5.3-Flash by 12.634.647.2+12.6
- AA IT-Bench SREGLM-5.3-Flash by 8.542.751.2+8.5
ReasoningGLM-5.2
- ARC-AGI-2GLM-5.3-Flash by 4322.865.8+43
- CritPtGLM-5.2 by 5.520.915.4+5.5
- GPQA DiamondGLM-5.3-Flash by 1.789.591.2+1.7
- Humanity's Last ExamGLM-5.2 by 1.241.139.9+1.2
- LiveBench · ReasoningGLM-5.2 by 178.677.6+1
FactsGLM-5.2
- AA-Omniscience · AccuracyGLM-5.3-Flash by 3.224.327.5+3.2
- SimpleQA VerifiedGLM-5.2 by 2.234.232+2.2
- AA-Omniscience · Non-hallucinationGLM-5.2 by 1.373.772.4+1.3
MathGLM-5.2
- FrontierMath Tier 4GLM-5.2 by 12.229.317.1+12.2
- LiveBench · MathematicsGLM-5.2 by 8.689.881.2+8.6
- FrontierMath Tiers 1-3 (v2)GLM-5.2 by 3.459.255.8+3.4
Following instructionsGLM-5.2
- LiveBench · Instruction FollowingGLM-5.2 by 9.562.352.8+9.5
Other results14 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.
- ToolathlonGLM-5.3-Flash by 30.248.278.4+30.2
- OfficeQA ProGLM-5.3-Flash by 2141.462.4+21
- DeepSWE 1.1GLM-5.3-Flash by 19.44463.4+19.4
- JobBenchGLM-5.3-Flash by 17.343.460.7+17.3
- ARC-AGI-1GLM-5.3-Flash by 147791+14
- AA Agentic IndexGLM-5.3-Flash by 11.839.451.2+11.8
- AA IntelligenceGLM-5.3-Flash by 8.133.741.8+8.1
- NL2RepoGLM-5.3-Flash by 7.448.956.3+7.4
- GDP (Surge AI)GLM-5.3-Flash by 510.415.4+5
- AA-OmniscienceGLM-5.3-Flash by 3.14.47.5+3.1
- Artificial Analysis Coding IndexGLM-5.3-Flash by 2.768.871.5+2.7
- livebench_data_analysisGLM-5.3-Flash by 2.773.776.4+2.7
- Agents' Last ExamGLM-5.3-Flash by 2.523.826.3+2.5
- livebench_languageGLM-5.3-Flash by 1.176.277.3+1.1
Questions people ask
Which is better, GLM-5.2 or GLM-5.3-Flash?
GLM-5.2 wins four of the seven areas where both have results: reasoning, facts, math and following instructions. GLM-5.3-Flash wins coding, agents and long documents, and costs 89% less.
Which is better for coding?
GLM-5.3-Flash. It wins 2 of the 5 coding tests both models report; GLM-5.2 wins none, and 3 are ties.
Which is cheaper?
GLM-5.2 costs $1.40 per million input tokens and $4.40 per million output tokens; GLM-5.3-Flash costs $0.15 and $0.50. That makes GLM-5.3-Flash about 89% cheaper for the same work.
How do you compare the two?
We use the 36 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 14 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.