BenchLeader

Claude Opus 4.8 vs GLM 5.2

Full profiles: Claude Opus 4.8 (Anthropic) and GLM 5.2 (Zhipu AI): every benchmark, price and rank.

Updated · Data from 13 sources

As of 16 Nov 2026, Claude Opus 4.8 and GLM 5.2 are neck and neck on quality (61.1 vs 58.9 on the BenchLeader Index, inside the ±5.6 margin), and GLM 5.2 is 4.7× cheaper at $2.15 vs $10.00 per 1M tokens blended. Choose Claude Opus 4.8 for knowledge (61 vs 44), maths (68 vs 53), and reasoning (64 vs 57); choose GLM 5.2 for instruction following (72 vs 62) and agents & tools (62 vs 57).

Key facts (as of 16 Nov 2026)

  • Claude Opus 4.8 scores 61.1 on the BenchLeader Index and GLM 5.2 scores 58.9 (BenchLeader, 16 Nov 2026).
  • Claude Opus 4.8 costs $10.00 per 1M tokens blended (list price, 16 Nov 2026).
  • GLM 5.2 costs $2.15 per 1M tokens blended (list price, 16 Nov 2026).
  • Of the 43 benchmarks both models have, Claude Opus 4.8 leads 37 (BenchLeader, 16 Nov 2026).

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingEither62 vs 62 category score; SciCode 53.5% vs 51.2%Either
Agents & tool useGLM 5.262 vs 57 category score; τ³-Banking (AA) 34.6% vs 34.2%Medium
MathsClaude Opus 4.868 vs 53 category score; FrontierMath Tiers 1–3 80.0% vs 59.2%High
ReasoningClaude Opus 4.864 vs 57 category score; GPQA Diamond 91.0% vs 91.9%Medium
KnowledgeClaude Opus 4.861 vs 44 category score; SimpleQA Verified 53.0% vs 34.2%High
Long documentsEither63 vs 64 category score; AA-LCR v1.1 77.7% vs 78.3%Either
Following instructionsGLM 5.272 vs 62 category score; IFBench 73.3% vs 62.2%High
Chat quality (human preference)–Only GLM 5.2 has an Arena rating (66 category score)Not comparable
Lowest costGLM 5.2$2.15 vs $10.00 per 1M blended; agentic coding session $0.102 vs $0.400–
Fastest first answerGLM 5.226.8 s (GLM 5.2 (max)) vs 39.1 s (Claude Opus 4.8 (max)) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • Claude Opus 4.8 (max) and GLM 5.2 (max) are neck and neck on quality (61.1 vs 58.9, within the ±5.6 margin).
  • Claude Opus 4.8 (max) is stronger in coding, knowledge, maths, reasoning.
  • GLM 5.2 (max) is stronger in agents & tools, instruction following, long context.
  • GLM 5.2 (max) is 4.7× cheaper ($2.15 vs $10.00 per 1M blended).

At a glance

SpecClaude Opus 4.8GLM 5.2
DeveloperAnthropicZhipu AI
Released28 May 202616 Jun 2026
Context window (tokens)1,000,0001,000,000
Compared atClaude Opus 4.8 (max)GLM 5.2 (max)
Open weightsNoYes
Price in / out / cached input per 1M$5.00 / $25.00 / $0.500$1.40 / $4.40 / $0.260

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryClaude Opus 4.8 (max)GLM 5.2 (max)Leader
Reasoning64.157.2Claude Opus 4.8
Coding61.961.8level
Agents & tools56.962.1GLM 5.2
Maths67.752.6Claude Opus 4.8
Knowledge61.044.3Claude Opus 4.8
Instruction following62.171.8GLM 5.2
Human preference–66.4not comparable
Long context63.463.8level

On one chart

Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.

2550 avg75ReasoningCodingAgents & toolsMathsKnowledgeInstructionfollowingLong contextClaude Opus 4.8, Reasoning: 64.1Claude Opus 4.8, Coding: 61.9Claude Opus 4.8, Agents & tools: 56.9Claude Opus 4.8, Maths: 67.7Claude Opus 4.8, Knowledge: 61.0Claude Opus 4.8, Instruction following: 62.1Claude Opus 4.8, Long context: 63.4GLM 5.2, Reasoning: 57.2GLM 5.2, Coding: 61.8GLM 5.2, Agents & tools: 62.1GLM 5.2, Maths: 52.6GLM 5.2, Knowledge: 44.3GLM 5.2, Instruction following: 71.8GLM 5.2, Long context: 63.8

Not drawn, measured for fewer than two of these models: Multimodal, Human preference.

Model, click to add or removeIndex

Where sources disagree

Across the 43 benchmarks both have, Claude Opus 4.8 leads 37, GLM 5.2 leads 5 and 1 is a tie. Every overall index that rates both puts Claude Opus 4.8 ahead.

Overall indexClaude Opus 4.8 (max)GLM 5.2 (max)Leader
BenchLeader Index61.158.9neck and neck
Arena Text1482†1475Claude Opus 4.8

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.

MeasureClaude Opus 4.8 (max)GLM 5.2 (max)
Blended price per 1M tokens$10.00$2.15
Input / output per 1M$5.00 / $25.00$1.40 / $4.40
Cached input per 1M$0.500$0.260
Price per task: chat reply$0.0095$0.0019
Price per task: summarize a 30-page report$0.075$0.019
Price per task: code edit$0.068$0.015
Price per task: agentic coding session$0.400$0.102
Price per task: structured extraction$0.015$0.0037
Agentic coding session, input mostly cached$0.198$0.050
Output speed85 tok/s69 tok/s
Time to first answer39.1 s26.8 s
Time to a complete answer, last measured 13 Nov 202681.0 s56.7 s

Each model is compared at its best-ranked reasoning-effort setting. Claude Opus 4.8 (max) scores 61.1 and takes 39 s to its first answer token; Claude Opus 4.8 (high) scores 60.4 in 1 s, Claude Opus 4.8 scores 58.7 in 1 s. GLM 5.2 (max) scores 58.9 and takes 27 s to its first answer token; GLM 5.2 scores 51.9 in 1 s, GLM 5.2 (no reasoning) scores 45.8 in 2 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

140 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherClaude Opus 4.8 (max)GLM 5.2 (max)
BenchLeader IndexBenchLeader61.158.9
Agents & tools scoreBenchLeader56.962.1
Coding scoreBenchLeader61.961.8
Instruction following scoreBenchLeader62.171.8
Knowledge scoreBenchLeader61.044.3
Long context scoreBenchLeader63.463.8
Maths scoreBenchLeader67.752.6
Reasoning scoreBenchLeader64.157.2
Human preference scoreBenchLeader–66.4
Blended price $/MBenchLeader$10.00$2.15
Output speedBenchLeader85 tok/s69 tok/s
Time to first answerBenchLeader39.1 s26.8 s
Context windowBenchLeader1,000,000 tokens1,000,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub91.0%91.9%
SimpleBenchSimpleBench64.8%†58.8%†
Arena Hard PromptsArena (LMArena)1505†1496
LiveBench 1.0 ReasoningLiveBench89.2%78.6%†
GPQA Diamond (AA)Artificial Analysis92.0%89.5%
Humanity's Last Exam (AA)Artificial Analysis48.7%41.1%
HLE-Diamond (AA)Artificial Analysis–17.7%
GPQA Diamond (Vals)Vals AI92.4%85.6%†
EnigmaEvalScale AI SEAL23.5%†–
Kagi LLM BenchmarkKagi LLM Benchmark88.8%†62.6%†
ARC-AGI-1ARC Prize92.5%77.0%†
ARC-AGI-2ARC Prize72.1%†22.8%†
ARC-AGI-3ARC Prize1.5%†–
CritPtArtificial Analysis20.9%20.9%
Arena ExpertArena (LMArena)1514†1502
Chess PuzzlesEpoch AI Benchmarking Hub52.0%22.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub36.0%18.0%†
LMCAEpoch AI Benchmarking Hub57.5%45.8%
DTBenchEpoch AI Benchmarking Hub94.9%93.6%
ForecastBenchForecastBench66.1%–
BTF-3Epoch AI Benchmarking Hub14.0%†–
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub–78.7%
SciCodeSciCode53.5%51.2%
WeirdMLWeirdML70.5%†70.1%
FrontierCodeCognition46.5%†24.5%†
Show all 140 benchmarks
MetricPublisherClaude Opus 4.8 (max)GLM 5.2 (max)
Coding
GSO-BenchGSO-Bench47.1%†–
Arena CodingArena (LMArena)1529†1515
Arena WebDevArena (LMArena)1536†1608
LiveBench 1.0 CodingLiveBench81.8%79.7%†
SciCode (AA)Artificial Analysis54.4%51.2%
LiveCodeBenchVals AI87.8%69.5%†
SWE-bench (Vals)Vals AI88.6%82.8%
Code MigrationVals AI47.3%37.9%
ProgramBenchVals AI1.0%0.5%
Vibe Code Bench v1.1Vals AI82.7%64.0%
SWE Atlas v1.0: Codebase QnAScale AI SEAL57.3%†48.1%†
SWE Atlas v1.0: RefactoringScale AI SEAL46.7%†42.4%†
SWE Atlas v1.0: Test WritingScale AI SEAL49.6%†41.5%†
MirrorCodeEpoch AI Benchmarking Hub65.6%36.7%
Surface Evolver BenchEpoch AI Benchmarking Hub87.5%†55.6%†
DeepSWE v1.1Epoch AI Benchmarking Hub59.0%43.8%
ALE-BenchEpoch AI Benchmarking Hub1563.8†1010.2
Terminal-Bench 4.0 (AA)Artificial Analysis21.7%1.0%
Terminal-Bench 2.1 (AA)Artificial Analysis84.6%77.9%
Startrise FrontendStartrise–72.1%†
GameBench 2 OverallDuelLab GameBench 249.4†43.1†
GameBench 2 by reasoning settingDuelLab GameBench 245.1†32.4†
Design Arena WebsiteDesign Arena1250†1277†
Agents & tools
Terminal-BenchTerminal-Bench23.6%–
OSWorld-Verified 2.0OSWorld20.6%–
Remote Labor IndexScale AI / CAIS8.3%†–
APEX-AgentsMercor48.9%45.2%†
Arena AgentArena (LMArena)–0.2
LiveBench Agentic CodingLiveBench50.5%51.8%†
Terminal-Bench HardArtificial Analysis58.3%50.8%
τ²-Bench Telecom (AA)Artificial Analysis94.4%99.1%
Terminal-Bench 2.1 (Vals)Vals AI71.9%67.8%
MCP AtlasScale AI SEAL82.2%77.8%†
HiL-BenchScale AI SEAL35.3%†43.7%†
τ³-bench Bankingτ²-bench39.7%37.1%†
GDPval-AA v2.1Artificial Analysis14611380
τ³-Banking (AA)Artificial Analysis34.2%34.6%
ITBench SRE (AA)Artificial Analysis–42.7%
Analyst Agent (AA)Artificial Analysis45.0%–
APEX-Agents (AA)Artificial Analysis–33.7%
Finance Agent v2Vals AI53.9%49.7%†
Harvey's Legal Agent Benchmark v1Vals AI9.6%7.1%
Legal Research BenchVals AI43.8%31.3%
SkillsBenchVals AI59.2%45.1%
SREBenchVals AI–0.0%
Tax Agent BenchVals AI30.0%–
Terminal-Bench 2.0 (Vals)Vals AI70.0%–
Terminal-Bench 4.0 (Vals)Vals AI23.2%–
Terminal-Bench ScienceVals AI4.3%–
Time Horizon Index: KSPVals AI11.8%–
Arena SearchArena (LMArena)1194†–
Arena Agent Task CompletionArena (LMArena)–1.7
DeepResearch BenchDeepResearch Bench50.2%†–
PostTrainBenchEpoch AI Benchmarking Hub31.0%30.4%
GBAEvalEpoch AI Benchmarking Hub70.9%†0.0%†
Vending-Bench 2Andon Labs2992.38313.8†
AutomationBenchArtificial Analysis45.6%28.4%
GDP.pdf (AA)Artificial Analysis22.8%10.4%
EnterpriseOps-GymArtificial Analysis44.0%42.7%
AA-Briefcase v1.1Artificial Analysis13201226
Design Arena AgenticDesign Arena1226†1194†
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub80.0%59.2%
FrontierMath Tier 4Epoch AI Benchmarking Hub56.1%29.3%
OTIS Mock AIMEEpoch AI Benchmarking Hub98.3%86.4%
ProofBenchVals AI69.0%35.0%
LiveBench MathematicsLiveBench94.3%89.8%†
AIME 2026MathArena100.0%90.0%
HMMT February 2026MathArena95.5%92.4%
MathArena ApexMathArena81.3%19.8%
Arena MathsArena (LMArena)1476†1487
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub53.0%34.2%
LiveBench 1.0 Data AnalysisLiveBench66.0%73.7%†
AA-Omniscience v1.1Artificial Analysis–0.6
MMLU-ProVals AI89.6%86.7%†
LegalBenchVals AI83.6%84.1%†
CorpFinVals AI66.7%66.1%†
TaxEvalVals AI75.6%73.3%†
Excel Modeling BenchmarkVals AI69.4%61.5%†
MedCodeVals AI53.2%40.8%†
MedScribeVals AI85.8%83.5%†
Public Benefits Bench v1.1Vals AI68.1%–
DrugDiscoveryBenchScale AI SEAL46.8%36.2%†
EBR-benchEpoch AI Benchmarking Hub28.6%9.5%
GDP.pdfEpoch AI Benchmarking Hub24.0%–
AA-Omniscience v1.1: accuracyArtificial Analysis–21.1%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis–74.1%
Instruction following
LiveBench LanguageLiveBench79.7%76.2%†
LiveBench Instruction FollowingLiveBench72.0%62.3%†
IFBenchArtificial Analysis62.2%73.3%
Arena Instruction FollowingArena (LMArena)1479†1470
Human preference
Arena TextArena (LMArena)1474†1475
EQ-Bench 4EQ-Bench1268†1208†
EQ-Bench Creative Writing v3EQ-Bench1840†1759†
Arena Creative WritingArena (LMArena)1464†1451
Arena Multi-turnArena (LMArena)1493†1477
Arena Longer QueriesArena (LMArena)1499†1486
Multimodal
Arena VisionArena (LMArena)1289†–
MMMU-Pro (Vals)Vals AI86.6%–
MortgageTaxVals AI69.9%–
Humanity's Sixth SenseScale AI SEAL23.5%–
GeoBenchGeoBench81.0%†–
Blueprint-Bench 2Epoch AI Benchmarking Hub14.5%†–
Furniture AssemblyEpoch AI Benchmarking Hub42.5%–
Long context
AA-LCR v1.1Artificial Analysis77.7%78.3%
Arena DocumentArena (LMArena)1476†–
MLCRArtificial Analysis19.4%12.8%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub159.9†153.0†
LiveBench 1.0LiveBench76.2%73.2%†
AA Intelligence Index v4.3.2Artificial Analysis41.833.7
AA Progress Index v1Artificial Analysis318.4302.5
Vals RSI IndexVals AI24.0%–
Vals Multimodal IndexVals AI70.9%–
Safety & honesty
FORTRESSScale AI SEAL18.2%48.7%†
DistressBenchScale AI SEAL71.4%†–
Metadata
AA Openness IndexArtificial Analysis–44.4

Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • Claude Opus 4.8 (max): BenchLeader Index 60.3 → 61.1.
  • Claude Opus 4.8 (max): 27 new results (LiveCodeBench, MMLU-Pro, LegalBench…).
  • Claude Opus 4.8 (max): 1 new result (GDPval-AA v2.1).
  • GLM 5.2 (max): 1 new result (GDPval-AA v2.1).
  • GLM 5.2 (max): BenchLeader Index 59.5 → 59.0.
  • Claude Opus 4.8 (max): 1 score revision.
  • GLM 5.2 (max): 3 new results (AIME 2026, HMMT February 2026, MathArena Apex); 1 score revision.
  • Claude Opus 4.8 (max): 1 new result (MirrorCode).
  • GLM 5.2 (max): 1 new result (MirrorCode).
  • GLM 5.2 (max): BenchLeader Index 60.2 → 59.5.

Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs

Claude Opus 4.8 vs GLM 5.2: questions

Is Claude Opus 4.8 better than GLM 5.2?
Claude Opus 4.8 (max) and GLM 5.2 (max) are neck and neck on quality (61.1 vs 58.9, within the ±5.6 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
Claude Opus 4.8 vs GLM 5.2: which is cheaper, and will that change?
GLM 5.2 is cheaper today: $2.15 against $10.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Claude Opus 4.8 or GLM 5.2 better for coding?
They are level in coding (62 vs 62 on the category index). On SciCode, the top shared coding benchmark, it is 53.5% for Claude Opus 4.8 against 51.2% for GLM 5.2.
Which is faster, Claude Opus 4.8 or GLM 5.2?
Claude Opus 4.8 streams faster (85 vs 69 output tokens per second, Artificial Analysis 7-day median).
Is Claude Opus 4.8 or GLM 5.2 better for agents and tool use?
GLM 5.2 scores higher in agentic tasks (62 vs 57 on the category index, where 50 is average). On τ³-Banking (AA), the top shared agents & tools benchmark, it is 34.2% for Claude Opus 4.8 against 34.6% for GLM 5.2.
More questions (4)
Which is better for long documents?
They are level on long-context benchmarks (63 vs 64). Both accept 1,000,000 tokens of context.
Which has lower latency?
GLM 5.2 answers first: 26.8 s against 39.1 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Do independent benchmarks agree on Claude Opus 4.8 vs GLM 5.2?
Across the 43 benchmarks both have, Claude Opus 4.8 leads 37, GLM 5.2 leads 5 and 1 is a tie. Every overall index that rates both puts Claude Opus 4.8 ahead.
When was GLM 5.2 released?
GLM 5.2 was released on 16 Jun 2026.