Claude Opus 4.6 vs GLM-5
Wins 5 of 7 areas
Coding · Agents · Reasoning · Math · Long documents
Wins 2 of 7 areas
Facts · Following instructions
Claude Opus 4.6 is the stronger all-rounder.GLM-5 is cheaper and better at facts.
Scores updated · 42 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.
- ReasoningHard problems that need careful thinking50Claude Opus 4.65 of 5 tests
- CodingWriting and fixing software41Claude Opus 4.64 of 5 tests
- AgentsCarrying out multi-step tasks on its own31Claude Opus 4.63 of 4 tests
- MathCompetition and research-level math31Claude Opus 4.63 of 4 tests
- Long documentsFinding answers in very long texts10Claude Opus 4.61 of 1 test
- FactsGetting facts right instead of making them up12GLM-52 of 3 tests
- Following instructionsDoing exactly what it is asked01GLM-51 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 costs 86% 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 Opus 4.6 pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+63.9points ahead
- Proof problems from the US Math OlympiadUSAMO 2026+31.1points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+20.7points ahead
Where GLM-5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+27.5points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+19.2points ahead
- Olympiad math problems with short answersIMOAnswerBench+7.2points ahead
Every test, side by side
All 42 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4.6
- LMArena · WebDevClaude Opus 4.6 by 112 rating points15461434+112 rating
- SciCodeClaude Opus 4.6 by 5.751.946.2+5.7
- SWE-bench VerifiedClaude Opus 4.6 by 380.877.8+3
- Terminal-Bench HardClaude Opus 4.6 by 346.243.2+3
- SWE-bench ProGLM-5 by 1.753.455.1+1.7
AgentsClaude Opus 4.6
- AA ApexAgentsClaude Opus 4.6 by 18.53314.5+18.5
- GDPValClaude Opus 4.6 by 11.355.944.6+11.3
- BrowseCompClaude Opus 4.6 by 8.18475.9+8.1
- MCP AtlasGLM-5 by 5.162.767.8+5.1
ReasoningClaude Opus 4.6
- ARC-AGI-2Claude Opus 4.6 by 63.968.84.9+63.9
- SimpleBenchClaude Opus 4.6 by 14.467.653.2+14.4
- CritPtClaude Opus 4.6 by 10.612.62+10.6
- Humanity's Last ExamClaude Opus 4.6 by 10.639.929.3+10.6
- GPQA DiamondClaude Opus 4.6 by 7.689.682+7.6
FactsGLM-5
- AA-Omniscience · Non-hallucinationGLM-5 by 27.537.264.7+27.5
- AA-Omniscience · AccuracyClaude Opus 4.6 by 20.74726.3+20.7
- Vectara HHEM hallucination ratelower is betterGLM-5 by 2.112.210.1+2.1
MathClaude Opus 4.6
- USAMO 2026Claude Opus 4.6 by 31.166.235.1+31.1
- HMMT Feb. 2026Claude Opus 4.6 by 13.496.282.8+13.4
- IMOAnswerBenchGLM-5 by 7.275.382.5+7.2
- AIME 2026Claude Opus 4.6 by 1.396.795.4+1.3
Other results19 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.
- ARC-AGI-1Claude Opus 4.6 by 48.39344.7+48.3
- CyberGymClaude Opus 4.6 by 25.573.848.3+25.5
- frontiermath_tier_4_v1Claude Opus 4.6 by 20.822.92.1+20.8
- NL2RepoClaude Opus 4.6 by 13.949.835.9+13.9
- AA-OmniscienceClaude Opus 4.6 by 13.413.70.3+13.4
- Terminal-Bench 2.0Claude Opus 4.6 by 9.265.456.2+9.2
- Tool-DecathlonClaude Opus 4.6 by 9.247.238+9.2
- browsecomp_with_context_managerClaude Opus 4.6 by 8.18475.9+8.1
- vectara_avg_summary_lengthClaude Opus 4.6 by 63.2 rating points137.674.4+63.2 rating
- τ²-Bench Telecom (AA run)GLM-5 by 6.192.198.2+6.1
- AA Agentic IndexClaude Opus 4.6 by 4.567.663.1+4.5
- AA IntelligenceClaude Opus 4.6 by 431.927.9+4
- Artificial Analysis Coding IndexClaude Opus 4.6 by 3.948.144.2+3.9
- τ³-BenchClaude Opus 4.6 by 3.272.469.2+3.2
- AIME 2025Claude Opus 4.6 by 3.199.896.7+3.1
- HLE (with tools)Claude Opus 4.6 by 2.953.350.4+2.9
- vectara_factual_consistencyGLM-5 by 2.187.889.9+2.1
- HMMT Nov. 2025tie96.396.9tie
- vectara_answer_ratetie99.899.7tie
Questions people ask
Which is better, Claude Opus 4.6 or GLM-5?
Claude Opus 4.6 wins five of the seven areas where both have results: coding, agents, reasoning, math and long documents. GLM-5 wins facts and following instructions, and costs 86% less.
Which is better for coding?
Claude Opus 4.6. It wins 4 of the 5 coding tests both models report; GLM-5 wins 1.
Which is cheaper?
Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens; GLM-5 costs $1.00 and $3.20. That makes GLM-5 about 86% cheaper for the same work.
How do you compare the two?
We use the 42 benchmark tests both models have published scores on. The verdict counts the 23 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 19 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.