Claude Sonnet 5 vs GLM-5.2
Wins 6 of 7 areas
Coding · Agents · Reasoning · Math · Long documents · Following instructions
Wins 0 of 7 areas
—
Claude Sonnet 5 is the stronger all-rounder.GLM-5.2 is cheaper.
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 software51Claude Sonnet 55 of 6 tests
- AgentsCarrying out multi-step tasks on its own30Claude Sonnet 53 of 3 tests
- ReasoningHard problems that need careful thinking31Claude Sonnet 53 of 5 tests · 1 tie
- MathCompetition and research-level math20Claude Sonnet 52 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts10Claude Sonnet 51 of 1 test
- Following instructionsDoing exactly what it is asked10Claude Sonnet 51 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.
GLM-5.2 costs 52% 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 Claude Sonnet 5 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+15.8points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+13.1points ahead
- Fresh logic puzzles that change regularlyLiveBench · Reasoning+10.1points ahead
Where GLM-5.2 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+13.1points ahead
- Research-level physics problemsCritPt+4points 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.
CodingClaude Sonnet 5
- LiveBench · Agentic CodingClaude Sonnet 5 by 7.659.451.8+7.6
- LMArena · WebDevGLM-5.2 by 66 rating points15391605+66 rating
- SciCodeClaude Sonnet 5 by 3.154.351.2+3.1
- Terminal-Bench 2.1Claude Sonnet 5 by 2.680.577.9+2.6
- SWE-bench ProClaude Sonnet 5 by 1.163.262.1+1.1
- LiveBench · CodingClaude Sonnet 5 by 180.779.7+1
AgentsClaude Sonnet 5
- Terminal-Bench 4.0Claude Sonnet 5 by 13.114.11+13.1
- GDPValClaude Sonnet 5 by 4.648.343.7+4.6
- τ-Bench V3 · BankingClaude Sonnet 5 by 2.737.334.6+2.7
ReasoningClaude Sonnet 5
- LiveBench · ReasoningClaude Sonnet 5 by 10.188.778.6+10.1
- CritPtGLM-5.2 by 416.920.9+4
- SimpleBenchClaude Sonnet 5 by 1.860.658.8+1.8
- GPQA DiamondClaude Sonnet 5 by 1.691.189.5+1.6
- Humanity's Last Examtie41.341.1tie
FactsEven
- AA-Omniscience · AccuracyClaude Sonnet 5 by 15.84024.3+15.8
- AA-Omniscience · Non-hallucinationGLM-5.2 by 13.160.673.7+13.1
- SimpleQA Verifiedtie33.734.2tie
MathClaude Sonnet 5
- FrontierMath Tiers 1-3 (v2)Claude Sonnet 5 by 6.465.659.2+6.4
- LiveBench · MathematicsClaude Sonnet 5 by 3.192.989.8+3.1
- FrontierMath Tier 4tie29.329.3tie
Following instructionsClaude Sonnet 5
- LiveBench · Instruction FollowingClaude Sonnet 5 by 1.663.962.3+1.6
Other results15 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.
- OfficeQA ProClaude Sonnet 5 by 1859.441.4+18
- Toolathlon VerifiedClaude Sonnet 5 by 14.874.759.9+14.8
- Program BenchClaude Sonnet 5 by 13.677.363.7+13.6
- AA-OmniscienceClaude Sonnet 5 by 12.116.44.4+12.1
- GDPval-AA v2Claude Sonnet 5 by 110 rating points16181508+110 rating
- DeepSWE 1.1Claude Sonnet 5 by 105444+10
- Terminal-bench 3.0Claude Sonnet 5 by 1014.64.6+10
- GDPval-AA v2 EloClaude Sonnet 5 by 74 rating points15841510+74 rating
- ToolathlonClaude Sonnet 5 by 6.154.348.2+6.1
- AA Agentic IndexClaude Sonnet 5 by 4.944.339.4+4.9
- AA IntelligenceClaude Sonnet 5 by 4.538.233.7+4.5
- Artificial Analysis Coding IndexClaude Sonnet 5 by 2.771.568.8+2.7
- HLE (with tools)Claude Sonnet 5 by 2.757.454.7+2.7
- livebench_data_analysisGLM-5.2 by 271.773.7+2
- livebench_languageGLM-5.2 by 1.27576.2+1.2
Questions people ask
Which is better, Claude Sonnet 5 or GLM-5.2?
Claude Sonnet 5 wins six of the seven areas where both have results: coding, agents, reasoning, math, long documents and following instructions. GLM-5.2 wins none, but costs 52% less. They are level on facts.
Which is better for coding?
Claude Sonnet 5. It wins 5 of the 6 coding tests both models report; GLM-5.2 wins 1.
Which is cheaper?
Claude Sonnet 5 costs $2.00 per million input tokens and $10.00 per million output tokens; GLM-5.2 costs $1.40 and $4.40. That makes GLM-5.2 about 52% 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 22 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 15 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.