BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases12

Claude Opus 5.5 vs GLM 5.2

Full profiles: Claude Opus 5.5 (Anthropic) and GLM 5.2 (Zhipu AI): every benchmark, price and rank.

Updated · Data from 12 sources

As of 9 Nov 2026, Claude Opus 5.5 is ahead of GLM 5.2 on quality (68.9 vs 59.4 on the BenchLeader Index), but GLM 5.2 is 3.7× cheaper at $2.15 vs $8.00 per 1M tokens blended. Claude Opus 5.5 leads in every category we measure both in.

Key facts (as of 9 Nov 2026)

  • Claude Opus 5.5 scores 68.9 on the BenchLeader Index and GLM 5.2 scores 59.4 (BenchLeader, 9 Nov 2026).
  • Claude Opus 5.5 costs $8.00 per 1M tokens blended (list price, 9 Nov 2026).
  • GLM 5.2 costs $2.15 per 1M tokens blended (list price, 9 Nov 2026).
  • Of the 27 benchmarks both models have, Claude Opus 5.5 leads 24 (BenchLeader, 9 Nov 2026).
  • On Arena WebDev, Claude Opus 5.5 is rated 1818 and GLM 5.2 1608 (LMArena, as published by 9 Nov 2026).
Claude Opus 5.5 benchmarks max Anthropic GLM 5.2 benchmarks max Zhipu AIAdd a third model

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
CodingClaude Opus 5.580 vs 62 category score; Arena WebDev 1818 vs 1608High
Agents & tool useEither63 vs 62 category score; Arena Agent 20 vs 0.4Either
MathsClaude Opus 5.574 vs 57 category score; FrontierMath Tiers 1–3 91.2% vs 59.2%High
ReasoningClaude Opus 5.572 vs 57 category score; HLE-Diamond (AA) 67.9% vs 17.7%High
KnowledgeClaude Opus 5.574 vs 45 category score; SimpleQA Verified 72.2% vs 34.2%High
Long documentsClaude Opus 5.567 vs 64 category score; AA-LCR v1.1 84.7% vs 78.3%Medium
Chat quality (human preference)–Only GLM 5.2 has an Arena rating (67 category score)Not comparable
Lowest costGLM 5.2$2.15 vs $8.00 per 1M blended; agentic coding session $0.102 vs $0.320–
Fastest first answerGLM 5.224.9 s (GLM 5.2 (max)) vs 380.6 s (Claude Opus 5.5 (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 5.5 (max) leads on quality: 68.9 vs 59.4.
  • Claude Opus 5.5 (max) is stronger in agents & tools, coding, knowledge, long context, maths, multimodal, reasoning.
  • GLM 5.2 (max) is stronger in human preference, instruction following.
  • GLM 5.2 (max) is 3.7× cheaper ($2.15 vs $8.00 per 1M blended).
  • Claude Opus 5.5 (max) streams 1.4× faster (115 vs 84 tokens per second).

At a glance

SpecClaude Opus 5.5GLM 5.2
DeveloperAnthropicZhipu AI
Released22 Sep 202616 Jun 2026
Context window (tokens)1,000,0001,000,000
Max output (tokens)128,000 (up to 300K on the Message Batches API (beta))–
Input typestext, image–
Reasoning effort settingslow, medium, high, xhigh, max–
Compared atClaude Opus 5.5 (max)GLM 5.2 (max)
Open weightsNoYes
Price in / out / cached input per 1M$4.00 / $20.00 / $0.200$1.40 / $4.40 / $0.260
API model IDclaude-opus-5-5–
Where to use itClaude API, Amazon Bedrock, Google Cloud, Microsoft Foundry–

Limits, availability and announced prices from the vendors' own pages: platform.claude.com (checked 20 Oct 2026). Everything else is from the day's data.

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 5.5 (max)GLM 5.2 (max)Leader
Reasoning72.357.4Claude Opus 5.5
Coding80.462.1Claude Opus 5.5
Agents & tools62.762.4level
Maths74.057.1Claude Opus 5.5
Knowledge73.744.5Claude Opus 5.5
Instruction following–71.8not comparable
Human preference–66.5not comparable
Multimodal70.3–not comparable
Long context67.163.8Claude Opus 5.5

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 & toolsMathsKnowledgeLong contextClaude Opus 5.5, Reasoning: 72.3Claude Opus 5.5, Coding: 80.4Claude Opus 5.5, Agents & tools: 62.7Claude Opus 5.5, Maths: 74.0Claude Opus 5.5, Knowledge: 73.7Claude Opus 5.5, Long context: 67.1GLM 5.2, Reasoning: 57.4GLM 5.2, Coding: 62.1GLM 5.2, Agents & tools: 62.4GLM 5.2, Maths: 57.1GLM 5.2, Knowledge: 44.5GLM 5.2, Long context: 63.8

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

ModelIndex

Where sources disagree

Across the 27 benchmarks both have, Claude Opus 5.5 leads 24, GLM 5.2 leads 3. Every overall index that rates both puts Claude Opus 5.5 ahead.

Overall indexClaude Opus 5.5 (max)GLM 5.2 (max)Leader
BenchLeader Index68.959.4Claude Opus 5.5
Arena Text1506†1475Claude Opus 5.5

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 5.5 (max)GLM 5.2 (max)
Blended price per 1M tokens$8.00$2.15
Input / output per 1M$4.00 / $20.00$1.40 / $4.40
Cached input per 1M$0.200$0.260
Price per task: chat reply$0.0076$0.0019
Price per task: summarize a 30-page report$0.060$0.019
Price per task: code edit$0.054$0.015
Price per task: agentic coding session$0.320$0.102
Price per task: structured extraction$0.012$0.0037
Agentic coding session, input mostly cached$0.149$0.050
Cost per benchmark task (AA Index)$10.04–
Output tokens per task (AA Index)173k–
Of which reasoning120k–
Time per task (AA Index)25 min–
Output speed115 tok/s84 tok/s
Time to first answer380.6 s24.9 s
Time to a complete answer771.1 s58.2 s

Each model is compared at its best-ranked reasoning-effort setting. Claude Opus 5.5 (max) scores 68.9 and takes 381 s to its first answer token; Claude Opus 5.5 (xhigh) scores 65.5 in 84 s, Claude Opus 5.5 (high) scores 65.4 in 28 s. GLM 5.2 (max) scores 59.4 and takes 25 s to its first answer token; GLM 5.2 scores 51.5 in 1 s, GLM 5.2 (no reasoning) scores 45.9 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

166 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 5.5 (max)GLM 5.2 (max)
BenchLeader IndexBenchLeader68.959.4
Agents & tools scoreBenchLeader62.762.4
Coding scoreBenchLeader80.462.1
Knowledge scoreBenchLeader73.744.5
Long context scoreBenchLeader67.163.8
Maths scoreBenchLeader74.057.1
Multimodal scoreBenchLeader70.3–
Reasoning scoreBenchLeader72.357.4
Human preference scoreBenchLeader–66.5
Instruction following scoreBenchLeader–71.8
Blended price $/MBenchLeader$8.00$2.15
Output speedBenchLeader115 tok/s84 tok/s
Time to first answerBenchLeader380.6 s24.9 s
Context windowBenchLeader1,000,000 tokens1,000,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub90.6%91.9%
SimpleBenchSimpleBench–58.8%†
Arena Hard PromptsArena (LMArena)1530†1496
LiveBench 2.0 ReasoningLiveBench92.2%†–
LiveBench 1.0 ReasoningLiveBench92.2%78.6%†
GPQA Diamond (AA)Artificial Analysis–89.5%
Humanity's Last Exam (AA)Artificial Analysis61.4%41.1%
HLE-Diamond (AA)Artificial Analysis67.9%17.7%
GPQA Diamond (Vals)Vals AI–85.6%†
Kagi LLM BenchmarkKagi LLM Benchmark–62.6%†
ARC-AGI-1ARC Prize97.5%77.0%†
ARC-AGI-2ARC Prize91.7%22.8%†
CritPtArtificial Analysis31.7%20.9%
CritPt v2 hard subset (AA)Artificial Analysis68.0%–
MysteryMechanismVals AI49.5%†–
Chess PuzzlesEpoch AI Benchmarking Hub67.0%22.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub71.0%18.0%†
LMCAEpoch AI Benchmarking Hub68.2%45.8%
DTBenchEpoch AI Benchmarking Hub98.9%93.6%
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub–78.7%
SciCodeSciCode66.9%51.2%
WeirdMLWeirdML–70.1%
FrontierCodeCognition54.6%†24.5%†
Arena CodingArena (LMArena)1546†1515
Arena WebDevArena (LMArena)18181608
Show all 166 benchmarks
MetricPublisherClaude Opus 5.5 (max)GLM 5.2 (max)
Coding
LiveBench 2.0 CodingLiveBench87.1%†–
LiveBench 1.0 CodingLiveBench89.3%79.7%†
LiveCodeBenchVals AI–69.5%†
IOIVals AI95.1%†–
SWE-bench (Vals)Vals AI–82.8%
Code MigrationVals AI66.7%†37.9%
Code Migration v1.1Vals AI81.0%†–
ProgramBenchVals AI18.5%†0.5%
SWE-bench MultimodalVals AI70.2%†–
Vibe Code Bench 1-100Vals AI30.4%†–
Vibe Code Bench v1.1Vals AI90.3%†64.0%
SWE Atlas: Codebase QnAScale AI SEAL–48.1%†
SWE Atlas: RefactoringScale AI SEAL–42.4%†
SWE Atlas: Test WritingScale AI SEAL–41.5%†
MirrorCodeEpoch AI Benchmarking Hub78.9%†–
Surface Evolver BenchEpoch AI Benchmarking Hub–55.6%†
DeepSWE v1.1Epoch AI Benchmarking Hub–43.8%
CursorBenchCursor57.8%–
ALE-BenchEpoch AI Benchmarking Hub2147.4†1010.2
FrontierSWEEpoch AI Benchmarking Hub62.3%–
Terminal-Bench 4.0 (AA)Artificial Analysis59.6%1.0%
Terminal-Bench 2.1 (AA)Artificial Analysis–77.9%
Startrise FrontendStartrise71.1%†72.1%†
GameBench 2 OverallDuelLab GameBench 287.6†43.2†
GameBench 2 by reasoning settingDuelLab GameBench 294.8†32.9†
Design Arena WebsiteDesign Arena1325†1281†
FrontierCode 1.1 (AA)Artificial Analysis52.2%–
FrontierCode 2.0 (AA)Artificial Analysis46.4%–
Agents & tools
Terminal-BenchTerminal-Bench65.8%–
Remote Labor IndexScale AI / CAIS21.3%†–
APEX-AgentsMercor73.5%45.2%†
Arena AgentArena (LMArena)200.4
LiveBench Long AgenticLiveBench72.8%†–
LiveBench BrowserLiveBench88.4%†–
LiveBench Agentic CodingLiveBench71.7%51.8%†
Terminal-Bench HardArtificial Analysis–50.8%
τ²-Bench Telecom (AA)Artificial Analysis–99.1%
Terminal-Bench 2.1 (Vals)Vals AI87.6%†67.8%
MCP AtlasScale AI SEAL85.4%†77.8%†
HiL-BenchScale AI SEAL59.0%†43.7%†
τ³-bench Bankingτ²-bench–37.1%†
GDPval-AA v2.1, normalised scoreArtificial Analysis68.3%–
GDPval-AA v2.1Artificial Analysis1866–
τ³-Banking (AA)Artificial Analysis–34.6%
ITBench SRE (AA)Artificial Analysis38.2%42.7%
Analyst Agent (AA)Artificial Analysis56.3%–
APEX-Agents (AA)Artificial Analysis–33.7%
CUA-benchVals AI14.0%–
CyberBenchVals AI55.4%†–
Finance Agent v2Vals AI58.6%†49.7%†
Finance Agent v2.1Vals AI67.8%†–
Harvey's Legal Agent Benchmark v1.1Vals AI9.2%†–
Harvey's Legal Agent Benchmark v1Vals AI3.8%†7.1%
Legal Research BenchVals AI50.5%†31.3%
Robo-benchVals AI14.6%–
SkillsBenchVals AI–45.1%
SREBenchVals AI33.6%†0.0%
Tax Agent BenchVals AI44.5%†–
Terminal-Bench 4.0 (Vals)Vals AI65.2%†–
Terminal-Bench ScienceVals AI47.1%†–
Time Horizon Index: KSPVals AI91.3%†–
Arena SearchArena (LMArena)1214†–
PostTrainBenchEpoch AI Benchmarking Hub–31.7%
GBAEvalEpoch AI Benchmarking Hub–0.0%†
Vending-Bench 2Andon Labs9235.3†8313.8†
AutomationBenchArtificial Analysis69.5%28.4%
GDP.pdf (AA)Artificial Analysis26.2%10.4%
Harvey LAB-AA v1.1Harvey4.2%–
EnterpriseOps-GymArtificial Analysis–42.7%
AA-Briefcase v1.1Artificial Analysis18001226
GDPxGDPx by SemiAnalysis12.0%–
Design Arena AgenticDesign Arena1334†1190†
Terminal-Bench-Science 0.1 (AA)Artificial Analysis59.0%–
CWE-Bench (AA)Artificial Analysis58.3%–
DeepSec-Bench (AA)Artificial Analysis26.9%–
CyberGym end-to-end (AA)Artificial Analysis0.8%–
FACTS SearchGoogle88.4%†–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub91.2%59.2%
FrontierMath Tier 4Epoch AI Benchmarking Hub95.0%29.3%
OTIS Mock AIMEEpoch AI Benchmarking Hub100.0%86.4%
ProofBenchVals AI100.0%35.0%
LiveBench MathematicsLiveBench97.1%89.8%†
AIME 2026MathArena–90.0%†
HMMT February 2026MathArena–92.4%†
MathArena ApexMathArena–19.8%†
Arena MathsArena (LMArena)1518†1487
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub72.2%34.2%
LiveBench 2.0 Data AnalysisLiveBench80.3%†–
LiveBench ResearchLiveBench88.9%†–
LiveBench 1.0 Data AnalysisLiveBench80.3%73.7%†
AA-Omniscience v1.1Artificial Analysis48.10.6
MMLU-ProVals AI–86.7%†
LegalBenchVals AI–84.1%†
CorpFinVals AI–66.1%†
TaxEvalVals AI–73.3%†
PRBench LegalScale AI SEAL47.6%†–
BioMysteryBenchVals AI79.3%†–
ClaimBenchVals AI100.0%†–
Excel Modeling BenchmarkVals AI75.9%†61.5%†
Excel Modeling Benchmark v1.1Vals AI75.8%†–
MedCodeVals AI49.8%†40.8%†
MedScribeVals AI91.4%†83.5%†
Public Benefits Bench v1.1Vals AI70.6%†–
DrugDiscoveryBenchScale AI SEAL–36.2%†
EBR-benchEpoch AI Benchmarking Hub71.4%9.5%
GDP.pdfEpoch AI Benchmarking Hub30.6%–
AA-Omniscience v1.1: accuracyArtificial Analysis65.5%21.1%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis49.7%74.1%
HealthBench Professional (AA)OpenAI59.9%–
Instruction following
LiveBench LanguageLiveBench86.3%76.2%†
LiveBench Instruction FollowingLiveBench65.7%62.3%†
IFBenchArtificial Analysis–73.3%
Arena Instruction FollowingArena (LMArena)1519†1470
Human preference
Arena TextArena (LMArena)1506†1475
EQ-Bench 4EQ-Bench1340†1208†
Arena Creative WritingArena (LMArena)1517†1451
Arena Multi-turnArena (LMArena)1500†1476
Arena Longer QueriesArena (LMArena)1520†1486
Multimodal
Arena VisionArena (LMArena)1324†–
LiveBench DesignLiveBench83.3%†–
MMMU-ProArtificial Analysis87.7%–
SAGEVals AI91.3%†–
Blueprint-Bench 2Epoch AI Benchmarking Hub51.2%†–
Code2Video-AAArtificial Analysis1536–
AA-3DArtificial Analysis1202–
Long context
AA-LCR v1.1Artificial Analysis84.7%78.3%
MLCRArtificial Analysis27.8%12.8%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub168.4†153.0†
LiveBench 2.0LiveBench83.4%†–
LiveBench 1.0LiveBench83.2%73.2%†
AA Intelligence Index v5.0Artificial Analysis59.1–
AA Intelligence Index v4.3.2Artificial Analysis57.633.7
AA Progress Index v1Artificial Analysis350302.5
Vals IndexVals AI70.1†–
Vals Finance IndexVals AI65.3%†–
Vals RSI IndexVals AI37.3%–
Vals Science IndexVals AI58.6%†–
Vals Router IndexVals AI64.2%†–
AA Cyber IndexArtificial Analysis28.7–
Safety & honesty
FORTRESSScale AI SEAL5.8%48.7%†
Metadata
AA Openness IndexArtificial Analysis–44.4

Data as of 2026-11-09. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261103-eaea31df94e6 (https://benchmarks.duellab.org/releases/gb2-20261103-eaea31df94e6/), 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 5.5 (max): 2 score revisions.
  • Claude Opus 5.5 (max): 1 score revision.
  • GLM 5.2 (max): BenchLeader Index 60.2 → 59.5.
  • GLM 5.2 (max): 4 new results (HLE-Diamond (AA), AA-Omniscience v1.1, AA-Omniscience v1.1: accuracy…).
  • Claude Opus 5.5 (max): 1 score revision.
  • Claude Opus 5.5 (max): 2 score revisions.
  • GLM 5.2 (max): cached input price moved from $0.30/M to $0.26/M (13% down).
  • Claude Opus 5.5 (max): 2 score revisions.
  • GLM 5.2 (max): 1 score revision; cached input price moved from $0.26/M to $0.30/M (15% up).
  • GLM 5.2 (max): cached input price moved from $0.30/M to $0.26/M (13% down).

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

Claude Opus 5.5 vs GLM 5.2: questions

Is Claude Opus 5.5 better than GLM 5.2?
Claude Opus 5.5 (max) leads on quality: 68.9 vs 59.4. The BenchLeader Index combines every independent quality benchmark; Claude Opus 5.5 (max) is ahead overall as of 9 Nov 2026, but check the category scores for your use.
Claude Opus 5.5 vs GLM 5.2: which is cheaper, and will that change?
GLM 5.2 is cheaper today: $2.15 against $8.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Claude Opus 5.5 or GLM 5.2 better for coding?
Claude Opus 5.5 scores higher in coding (80 vs 62 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1818 for Claude Opus 5.5 against 1608 for GLM 5.2.
Which is faster, Claude Opus 5.5 or GLM 5.2?
Claude Opus 5.5 streams faster (115 vs 84 output tokens per second, Artificial Analysis 7-day median).
Is Claude Opus 5.5 or GLM 5.2 better for agents and tool use?
They are level in agentic tasks (63 vs 62 on the category index). On Arena Agent, the top shared agents & tools benchmark, it is 20 for Claude Opus 5.5 against 0.4 for GLM 5.2.
More questions (4)
Which is better for long documents?
Claude Opus 5.5 scores higher on long-context benchmarks (67 vs 64). Both accept 1,000,000 tokens of context.
Which has lower latency?
GLM 5.2 answers first: 24.9 s against 380.6 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 5.5 vs GLM 5.2?
Across the 27 benchmarks both have, Claude Opus 5.5 leads 24, GLM 5.2 leads 3. Every overall index that rates both puts Claude Opus 5.5 ahead.
When was Claude Opus 5.5 released?
Claude Opus 5.5 was released on 22 Sep 2026.