Claude Sonnet 4.6 vs GLM-5.1
Wins 3 of 6 areas
Coding · Agents · Long documents
Wins 2 of 6 areas
Facts · Following instructions
Claude Sonnet 4.6 wins more areas, narrowly.GLM-5.1 is cheaper and better at facts.
Scores updated · 41 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 software50Claude Sonnet 4.65 of 6 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own32Claude Sonnet 4.63 of 6 tests · 1 tie
- Long documentsFinding answers in very long texts10Claude Sonnet 4.61 of 1 test
- FactsGetting facts right instead of making them up12GLM-5.12 of 3 tests
- Following instructionsDoing exactly what it is asked01GLM-5.11 of 1 test
- ReasoningHard problems that need careful thinking11Even1 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-5.1 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 Claude Sonnet 4.6 pulls ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+20.8points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+17.2points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+9.8points ahead
Where GLM-5.1 pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+19.7points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+18.5points ahead
- Multi-step tasks using many real software toolsMCP Atlas+10.5points ahead
Every test, side by side
All 41 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 4.6
- Terminal-Bench HardClaude Sonnet 4.6 by 9.85343.2+9.8
- Terminal-Bench 2.1Claude Sonnet 4.6 by 9.471.261.8+9.4
- SciCodeClaude Sonnet 4.6 by 5.350.144.8+5.3
- SWE-bench VerifiedClaude Sonnet 4.6 by 3.479.676.2+3.4
- LMArena · WebDevClaude Sonnet 4.6 by 12 rating points15211509+12 rating
- SWE-bench Protie58.158.4tie
AgentsClaude Sonnet 4.6
- τ-Bench V3 · BankingClaude Sonnet 4.6 by 20.834.413.6+20.8
- MCP AtlasGLM-5.1 by 10.561.371.8+10.5
- GDPValClaude Sonnet 4.6 by 5.736.731+5.7
- BrowseCompGLM-5.1 by 4.674.779.3+4.6
- Terminal-Bench 4.0Claude Sonnet 4.6 by 132+1
- AA IT-Bench SREtie39.840.3tie
ReasoningEven
- Humanity's Last ExamClaude Sonnet 4.6 by 3.533.630.1+3.5
- CritPtGLM-5.1 by 1.53.14.6+1.5
- GPQA Diamondtie87.586.8tie
FactsGLM-5.1
- AA-Omniscience · Non-hallucinationGLM-5.1 by 18.551.670.1+18.5
- AA-Omniscience · AccuracyClaude Sonnet 4.6 by 17.240.923.7+17.2
- SimpleQA VerifiedGLM-5.1 by 1.232.834+1.2
Other results21 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-5.1 by 2275.797.7+22
- GDPval-AA (Elo)Claude Sonnet 4.6 by 141 rating points16761535+141 rating
- AgentWorldBench - SWEClaude Sonnet 4.6 by 12.464.552.1+12.4
- AA-OmniscienceClaude Sonnet 4.6 by 11.312.20.8+11.3
- Terminal-Bench 2.0GLM-5.1 by 9.959.169+9.9
- AgentWorldBench - TerminalClaude Sonnet 4.6 by 9.75747.3+9.7
- AA Agentic IndexClaude Sonnet 4.6 by 7.933.125.2+7.9
- Artificial Analysis Coding IndexClaude Sonnet 4.6 by 7.26355.8+7.2
- AgentWorldBench - SearchClaude Sonnet 4.6 by 6.328.822.5+6.3
- Finance Agent v2Claude Sonnet 4.6 by 6.25144.8+6.2
- HLE (with tools)GLM-5.1 by 5.546.852.3+5.5
- LiveBenchClaude Sonnet 4.6 by 5.375.570.2+5.3
- AgentWorldBench - OverallClaude Sonnet 4.6 by 4.75651.3+4.7
- frontiermath_tier_4_v1GLM-5.1 by 4.28.312.5+4.2
- AgentWorldBench - OSClaude Sonnet 4.6 by 4.163.259.1+4.1
- AA IntelligenceClaude Sonnet 4.6 by 430.126.1+4
- GPQA (unspecified)Claude Sonnet 4.6 by 3.889.986.1+3.8
- AgentWorldBench - MCPClaude Sonnet 4.6 by 2.47067.6+2.4
- AgentWorldBench - AndroidGLM-5.1 by 1.15859.1+1.1
- AgentWorldBench - Webtie50.851.5tie
- Toolathlontie4140.7tie
Questions people ask
Which is better, Claude Sonnet 4.6 or GLM-5.1?
Claude Sonnet 4.6 wins three of the six areas where both have results: coding, agents and long documents. GLM-5.1 wins facts and following instructions, and costs 68% less. They are level on reasoning.
Which is better for coding?
Claude Sonnet 4.6. It wins 5 of the 6 coding tests both models report; GLM-5.1 wins none, 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-5.1 costs $1.38 and $4.40. That makes GLM-5.1 about 68% cheaper for the same work.
How do you compare the two?
We use the 41 benchmark tests both models have published scores on. The verdict counts the 20 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 21 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.