Claude Sonnet 4.6 vs GLM-5.2
Wins 1 of 7 areas
Long documents
Wins 6 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Following instructions
GLM-5.2 is the stronger all-rounder.Claude Sonnet 4.6 is better at long documents.
Scores updated · 38 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 software15GLM-5.25 of 7 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own14GLM-5.24 of 6 tests · 1 tie
- ReasoningHard problems that need careful thinking23GLM-5.23 of 5 tests
- FactsGetting facts right instead of making them up12GLM-5.22 of 3 tests
- Following instructionsDoing exactly what it is asked01GLM-5.21 of 2 tests · 1 tie
- MathCompetition and research-level math01GLM-5.21 of 1 test
- Long documentsFinding answers in very long texts10Claude Sonnet 4.61 of 1 test
Math and long documents 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 68% 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
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+22.1points ahead
- Research-level physics problemsCritPt+17.8points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+16.7points ahead
Where Claude Sonnet 4.6 pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+35.5points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+16.6points ahead
- Fresh logic puzzles that change regularlyLiveBench · Reasoning+6.2points ahead
Every test, side by side
All 38 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGLM-5.2
- LiveBench · Agentic CodingGLM-5.2 by 9.242.651.8+9.2
- LMArena · WebDevGLM-5.2 by 84 rating points15211605+84 rating
- Terminal-Bench 2.1GLM-5.2 by 6.771.277.9+6.7
- SWE-bench ProGLM-5.2 by 458.162.1+4
- Terminal-Bench HardClaude Sonnet 4.6 by 2.25350.8+2.2
- SciCodeGLM-5.2 by 1.150.151.2+1.1
- LiveBench · Codingtie79.379.7tie
AgentsGLM-5.2
- MCP AtlasGLM-5.2 by 15.561.376.8+15.5
- GDPValGLM-5.2 by 736.743.7+7
- AA ApexAgentsGLM-5.2 by 5.72833.7+5.7
- AA IT-Bench SREGLM-5.2 by 2.939.842.7+2.9
- Terminal-Bench 4.0Claude Sonnet 4.6 by 231+2
- τ-Bench V3 · Bankingtie34.434.6tie
ReasoningGLM-5.2
- ARC-AGI-2Claude Sonnet 4.6 by 35.558.322.8+35.5
- CritPtGLM-5.2 by 17.83.120.9+17.8
- Humanity's Last ExamGLM-5.2 by 7.533.641.1+7.5
- LiveBench · ReasoningClaude Sonnet 4.6 by 6.284.878.6+6.2
- GPQA DiamondGLM-5.2 by 287.589.5+2
FactsGLM-5.2
- AA-Omniscience · Non-hallucinationGLM-5.2 by 22.151.673.7+22.1
- AA-Omniscience · AccuracyClaude Sonnet 4.6 by 16.640.924.3+16.6
- SimpleQA VerifiedGLM-5.2 by 1.432.834.2+1.4
Following instructionsGLM-5.2
- IFBenchGLM-5.2 by 16.756.673.3+16.7
- LiveBench · Instruction Followingtie63.262.3tie
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.
- τ²-Bench Telecom (AA run)GLM-5.2 by 23.475.799.1+23.4
- DeepSWE 1.1GLM-5.2 by 143044+14
- OfficeQA ProClaude Sonnet 4.6 by 1253.441.4+12
- GDPval-AA v2GLM-5.2 by 113 rating points13951508+113 rating
- ARC-AGI-1Claude Sonnet 4.6 by 98677+9
- HLE (with tools)GLM-5.2 by 7.946.854.7+7.9
- AA-OmniscienceClaude Sonnet 4.6 by 7.812.24.4+7.8
- ToolathlonGLM-5.2 by 7.24148.2+7.2
- AA Agentic IndexGLM-5.2 by 6.333.139.4+6.3
- Artificial Analysis Coding IndexGLM-5.2 by 5.86368.8+5.8
- livebench_data_analysisClaude Sonnet 4.6 by 4.37873.7+4.3
- AA IntelligenceGLM-5.2 by 3.630.133.7+3.6
- livebench_languagetie76.176.2tie
Questions people ask
Which is better, Claude Sonnet 4.6 or GLM-5.2?
GLM-5.2 wins six of the seven areas where both have results: coding, agents, reasoning, facts, math and following instructions. Claude Sonnet 4.6 wins long documents.
Which is better for coding?
GLM-5.2. It wins 5 of the 7 coding tests both models report; Claude Sonnet 4.6 wins 1, and 1 is a tie.
Which is cheaper?
Claude Sonnet 4.6 costs $3.00 per million input tokens and $15.00 per million output tokens; GLM-5.2 costs $1.40 and $4.40. That makes GLM-5.2 about 68% cheaper for the same work.
How do you compare the two?
We use the 38 benchmark tests both models have published scores on. The verdict counts the 25 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.