Claude Opus 5 vs GLM 5.2
Full profiles: Claude Opus 5 (Anthropic) and GLM 5.2 (Zhipu AI): every benchmark, price and rank.
Updated · Data from 11 sources
As of 29 Oct 2026, Claude Opus 5 is ahead of GLM 5.2 on quality (65.5 vs 60.2 on the BenchLeader Index), but GLM 5.2 is 4.7× cheaper at $2.15 vs $10.00 per 1M tokens blended. Claude Opus 5 leads in every category we measure both in. GLM 5.2 streams ~1.5× faster.
Key facts (as of 29 Oct 2026)
- Claude Opus 5 scores 65.5 on the BenchLeader Index and GLM 5.2 scores 60.2 (BenchLeader, 29 Oct 2026).
- Claude Opus 5 costs $10.00 per 1M tokens blended (list price, 29 Oct 2026).
- GLM 5.2 costs $2.15 per 1M tokens blended (list price, 29 Oct 2026).
- GLM 5.2 streams ~1.5× faster than Claude Opus 5 (Artificial Analysis, 7-day median, 29 Oct 2026).
- Of the 35 benchmarks both models have, Claude Opus 5 leads 34 (BenchLeader, 29 Oct 2026).
- On Arena WebDev, Claude Opus 5 is rated 1690 and GLM 5.2 1606 (LMArena, as published by 29 Oct 2026).
Claude Opus 5 benchmarks 
GLM 5.2 benchmarks 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 case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Claude Opus 5 | 70 vs 62 category score; Arena WebDev 1690 vs 1606 | High |
| Agents & tool use | Claude Opus 5 | 64 vs 63 category score; τ³-Banking (AA) 42.1% vs 34.6% | Medium |
| Maths | Claude Opus 5 | 70 vs 57 category score; FrontierMath Tiers 1–3 85.6% vs 59.2% | High |
| Reasoning | Claude Opus 5 | 69 vs 66 category score; GPQA Diamond 93.9% vs 91.9% | Medium |
| Knowledge | Claude Opus 5 | 65 vs 44 category score; SimpleQA Verified 59.9% vs 34.2% | High |
| Long documents | Either | 65 vs 64 category score; AA-LCR v1.1 79.3% vs 78.3% | Either |
| Chat quality (human preference) | Claude Opus 5 | 68 vs 67 on Arena votes | Medium |
| Lowest cost | GLM 5.2 | $2.15 vs $10.00 per 1M blended; agentic coding session $0.102 vs $0.400 | – |
| Fastest first answer | GLM 5.2 | 24.1 s (GLM 5.2 (max)) vs 43.9 s (Claude Opus 5 (max)) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Claude Opus 5 (max) leads on quality: 65.5 vs 60.2.
- Claude Opus 5 (max) is stronger in agents & tools, coding, human preference, knowledge, long context, maths, multimodal, reasoning.
- GLM 5.2 (max) is stronger in instruction following.
- GLM 5.2 (max) is 4.7× cheaper ($2.15 vs $10.00 per 1M blended).
- GLM 5.2 (max) streams 1.7× faster (94 vs 56 tokens per second).
At a glance
| Spec | Claude Opus 5 | GLM 5.2 |
|---|---|---|
| Developer | Anthropic | Zhipu AI |
| Released | 24 Jul 2026 | 16 Jun 2026 |
| Context window (tokens) | 1,000,000 | 1,000,000 |
| Compared at | Claude Opus 5 (max) | GLM 5.2 (max) |
| Open weights | No | Yes |
| Price in / out / cached input per 1M | $5.00 / $25.00 / $0.500 | $1.40 / $4.40 / $0.299 |
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.
| Category | Claude Opus 5 (max) | GLM 5.2 (max) | Leader |
|---|---|---|---|
| Reasoning | 69.2 | 65.9 | Claude Opus 5 |
| Coding | 70.0 | 62.1 | Claude Opus 5 |
| Agents & tools | 64.3 | 62.8 | Claude Opus 5 |
| Maths | 70.4 | 57.2 | Claude Opus 5 |
| Knowledge | 65.0 | 43.6 | Claude Opus 5 |
| Instruction following | – | 71.7 | not comparable |
| Human preference | 68.4 | 66.5 | Claude Opus 5 |
| Multimodal | 67.6 | – | not comparable |
| Long context | 64.5 | 64.0 | level |
Where sources disagree
Across the 35 benchmarks both have, Claude Opus 5 leads 34, GLM 5.2 leads 1. Every overall index that rates both puts Claude Opus 5 ahead.
| Overall index | Claude Opus 5 (max) | GLM 5.2 (max) | Leader |
|---|---|---|---|
| BenchLeader Index | 65.5 | 60.2 | Claude Opus 5 |
| Arena Text | 1491 | 1475 | Claude Opus 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; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.
| Measure | Claude Opus 5 (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 | $0.299 |
| 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 |
| Output speed | 56 tok/s | 94 tok/s |
| Time to first answer | 43.9 s | 24.1 s |
| Time to a complete answer | 52.8 s | 29.5 s |
Each model is compared at its best-ranked reasoning-effort setting. Claude Opus 5 (max) scores 65.5 and takes 44 s to its first answer token; Claude Opus 5 scores 65.2 in 3 s, Claude Opus 5 (high) scores 64.4 in 14 s. GLM 5.2 (max) scores 60.2 and takes 24 s to its first answer token; GLM 5.2 scores 51.6 in 1 s, GLM 5.2 (no reasoning) scores 46.0 in 1 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
131 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.
| Metric | Publisher | Claude Opus 5 (max) | GLM 5.2 (max) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 65.5 | 60.2 |
| Agents & tools score | BenchLeader | 64.3 | 62.8 |
| Coding score | BenchLeader | 70.0 | 62.1 |
| Human preference score | BenchLeader | 68.4 | 66.5 |
| Knowledge score | BenchLeader | 65.0 | 43.6 |
| Long context score | BenchLeader | 64.5 | 64.0 |
| Maths score | BenchLeader | 70.4 | 57.2 |
| Multimodal score | BenchLeader | 67.6 | – |
| Reasoning score | BenchLeader | 69.2 | 65.9 |
| Instruction following score | BenchLeader | – | 71.7 |
| Blended price $/M | BenchLeader | $10.00 | $2.15 |
| Output speed | BenchLeader | 56 tok/s | 94 tok/s |
| Time to first answer | BenchLeader | 43.9 s | 24.1 s |
| Context window | BenchLeader | 1,000,000 tokens | 1,000,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 93.9% | 91.9% |
| SimpleBench | SimpleBench | 80.6%† | 58.8%† |
| Arena Hard Prompts | Arena (LMArena) | 1515 | 1496 |
| LiveBench 1.0 Reasoning | LiveBench | 91.2% | 78.6%† |
| GPQA Diamond (AA) | Artificial Analysis | 93.2% | 89.5% |
| Humanity's Last Exam (AA) | Artificial Analysis | 54.9% | 41.1% |
| GPQA Diamond (Vals) | Vals AI | 93.4%† | 85.6%† |
| Kagi LLM Benchmark | Kagi LLM Benchmark | – | 62.6%† |
| ARC-AGI-1 | ARC Prize | 97.5% | 77.0%† |
| ARC-AGI-2 | ARC Prize | 90.4% | 22.8%† |
| ARC-AGI-3 | ARC Prize | 30.2%† | – |
| CritPt | Artificial Analysis | 29.1% | 20.9% |
| MysteryMechanism | Vals AI | 37.4%† | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | 43.0% | 22.0% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 59.0% | 18.0%† |
| LMCA | Epoch AI Benchmarking Hub | 63.3% | 45.8% |
| DTBench | Epoch AI Benchmarking Hub | 97.6% | 93.6% |
| ForecastBench | ForecastBench | 66.5% | – |
| BTF-3 | Epoch AI Benchmarking Hub | 12.0%† | – |
| Coding | |||
| SWE-bench Verified (Epoch) | Epoch AI Benchmarking Hub | – | 78.7% |
| SciCode | SciCode | 56.4% | 51.2% |
| WeirdML | WeirdML | 91.8% | 70.1% |
| FrontierCode | Cognition | 53.4% | 24.5%† |
| Arena Coding | Arena (LMArena) | 1528 | 1514 |
| Arena WebDev | Arena (LMArena) | 1690 | 1606 |
Show all 131 benchmarks
| Metric | Publisher | Claude Opus 5 (max) | GLM 5.2 (max) |
|---|---|---|---|
| Coding | |||
| LiveBench 1.0 Coding | LiveBench | 81.5% | 79.7%† |
| LiveCodeBench | Vals AI | 89.0%† | 69.5%† |
| IOI | Vals AI | 84.3%† | – |
| SWE-bench (Vals) | Vals AI | 97.0%† | 82.8% |
| Code Migration | Vals AI | 57.5%† | 37.9% |
| ProgramBench | Vals AI | 3.0%† | 0.5% |
| Vibe Code Bench 1-100 | Vals AI | 28.5%† | – |
| Vibe Code Bench v1.1 | Vals AI | 88.4%† | 64.0% |
| SWE Atlas: Codebase QnA | Scale AI SEAL | 63.2%† | 48.1%† |
| SWE Atlas: Refactoring | Scale AI SEAL | – | 42.4%† |
| SWE Atlas: Test Writing | Scale AI SEAL | 62.2%† | 41.5%† |
| Surface Evolver Bench | Epoch AI Benchmarking Hub | – | 55.6%† |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | 73.7% | 43.8% |
| CursorBench | Cursor | 46.6% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 2164.6† | 1010.2 |
| FrontierSWE | Epoch AI Benchmarking Hub | 52.0% | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 49.0% | 1.0% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 89.1% | 77.9% |
| Startrise Frontend | Startrise | 82.3%† | 72.1%† |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 51.8% | – |
| OSWorld-Verified 2.0 | OSWorld | 31.4% | – |
| APEX-Agents | Mercor | 65.8% | 45.2%† |
| Arena Agent | Arena (LMArena) | 8.2 | 1.7 |
| LiveBench Agentic Coding | LiveBench | 65.2% | 51.8%† |
| Terminal-Bench Hard | Artificial Analysis | – | 50.8% |
| τ²-Bench Telecom (AA) | Artificial Analysis | – | 99.1% |
| Terminal-Bench 2.1 (Vals) | Vals AI | 84.6%† | 67.8% |
| MCP Atlas | Scale AI SEAL | 85.8%† | 77.8%† |
| HiL-Bench | Scale AI SEAL | 57.0%† | 43.7%† |
| τ³-bench Banking | τ²-bench | 48.7% | 37.1%† |
| τ³-Banking (AA) | Artificial Analysis | 42.1% | 34.6% |
| ITBench SRE (AA) | Artificial Analysis | – | 42.7% |
| Analyst Agent (AA) | Artificial Analysis | 53.8% | – |
| APEX-Agents (AA) | Artificial Analysis | – | 33.7% |
| CUA-bench | Vals AI | 9.0% | – |
| CyberBench | Vals AI | 65.4%† | – |
| Finance Agent v2 | Vals AI | 58.6%† | 49.7%† |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 6.7%† | 7.1% |
| Legal Research Bench | Vals AI | 55.3%† | 31.3% |
| SkillsBench | Vals AI | 60.4%† | 45.1% |
| SREBench | Vals AI | 12.2%† | 0.0% |
| Tax Agent Bench | Vals AI | 45.5%† | – |
| Terminal-Bench 4.0 (Vals) | Vals AI | 53.5%† | – |
| Terminal-Bench Science | Vals AI | 27.1%† | – |
| Time Horizon Index: KSP | Vals AI | 18.8%† | – |
| Arena Search | Arena (LMArena) | 1214† | – |
| BALROG | BALROG | 63.4% | – |
| PostTrainBench | Epoch AI Benchmarking Hub | 35.0%† | 31.7% |
| GBAEval | Epoch AI Benchmarking Hub | 79.6%† | 0.0%† |
| Vending-Bench 2 | Andon Labs | 11181.9† | 8313.8† |
| AutomationBench | Artificial Analysis | 56.6% | 28.4% |
| GDP.pdf (AA) | Artificial Analysis | 21.6% | 10.4% |
| EnterpriseOps-Gym | Artificial Analysis | 47.5% | 42.7% |
| AA-Briefcase v1.1 | Artificial Analysis | 1655 | 1227 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 28.6% | – |
| DeepSec-Bench (AA) | Artificial Analysis | 25.9% | – |
| FACTS Search | 85.3%† | – | |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 85.6% | 59.2% |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 73.2% | 29.3% |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 98.9% | 86.4% |
| ProofBench | Vals AI | 99.0% | 35.0% |
| LiveBench Mathematics | LiveBench | 95.7% | 89.8%† |
| AIME 2026 | MathArena | – | 90.0%† |
| HMMT February 2026 | MathArena | – | 92.4%† |
| MathArena Apex | MathArena | – | 19.8%† |
| Arena Maths | Arena (LMArena) | 1517 | 1488 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 59.9% | 34.2% |
| LiveBench 1.0 Data Analysis | LiveBench | 74.5% | 73.7%† |
| MMLU-Pro | Vals AI | 91.6%† | 86.7%† |
| LegalBench | Vals AI | 87.0%† | 84.1%† |
| CorpFin | Vals AI | 73.2%† | 66.1%† |
| TaxEval | Vals AI | 75.1%† | 73.3%† |
| BioMysteryBench | Vals AI | 79.3%† | – |
| Excel Modeling Benchmark | Vals AI | 73.6%† | 61.5%† |
| MedCode | Vals AI | 63.6%† | 40.8%† |
| MedScribe | Vals AI | 91.0%† | 83.5%† |
| Public Benefits Bench v1.1 | Vals AI | 76.9%† | – |
| DrugDiscoveryBench | Scale AI SEAL | 53.7% | 36.2%† |
| EBR-bench | Epoch AI Benchmarking Hub | 45.7% | 9.5% |
| GDP.pdf | Epoch AI Benchmarking Hub | 24.0% | – |
| Instruction following | |||
| LiveBench Language | LiveBench | 88.7% | 76.2%† |
| LiveBench Instruction Following | LiveBench | 63.8% | 62.3%† |
| IFBench | Artificial Analysis | – | 73.3% |
| Arena Instruction Following | Arena (LMArena) | 1495 | 1469 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1491 | 1475 |
| EQ-Bench 4 | EQ-Bench | 1365† | 1208† |
| Arena Creative Writing | Arena (LMArena) | 1474 | 1453 |
| Arena Multi-turn | Arena (LMArena) | 1487 | 1475 |
| Arena Longer Queries | Arena (LMArena) | 1503 | 1486 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1320† | – |
| MMMU-Pro | Artificial Analysis | 84.7% | – |
| MMMU-Pro (Vals) | Vals AI | 89.9%† | – |
| MortgageTax | Vals AI | 72.1%† | – |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 30.4%† | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 79.3% | 78.3% |
| Arena Document | Arena (LMArena) | 1490† | – |
| MLCR | Artificial Analysis | 11.1% | 12.8% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 163.9† | 151.9† |
| LiveBench 1.0 | LiveBench | 80.1% | 73.2%† |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 50.8 | 33.7 |
| AA Progress Index v1 | Artificial Analysis | 333.7 | 302.5 |
| Vals RSI Index | Vals AI | 33.0% | – |
| Vals Science Index | Vals AI | 47.9%† | – |
| Vals Multimodal Index | Vals AI | 73.9%† | – |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 16.8%† | – |
| Metadata | |||
| AA Openness Index | Artificial Analysis | – | 44.4 |
Data as of 2026-10-29. Best configuration of each model; every score links to its source on the model pages.
What changed
- GLM 5.2 (max): cached input price moved from $0.26/M to $0.30/M (15% up).
- Claude Opus 5 (max): BenchLeader Index 66.3 → 65.5.
- GLM 5.2 (max): BenchLeader Index 61.3 → 60.2.
- Claude Opus 5 (max): 1 new result (AA Progress Index v1).
- GLM 5.2 (max): 1 new result (AA Progress Index v1); cached input price moved from $0.30/M to $0.26/M (13% down).
- Claude Opus 5 (max): BenchLeader Index 66.8 → 66.3.
- GLM 5.2 (max): BenchLeader Index 61.8 → 61.3.
- GLM 5.2 (max): cached input price moved from $0.26/M to $0.30/M (15% up).
- GLM 5.2 (max): cost per task moved from $1.47/M to $1.28/M (13% down); cached input price moved from $0.30/M to $0.26/M (13% down).
- GLM 5.2 (max): 1 new result (AA-Briefcase v1.1); cached input price moved from $0.26/M to $0.30/M (15% up).
Related comparisons
- Claude Opus 5 vs Claude Opus 5.5
- Claude Opus 5 vs GPT-6 Astra
- Claude Opus 5 vs GPT-6.1 Sol
- GLM 5.2 vs Claude Opus 5.5
- GLM 5.2 vs GPT-6 Astra
- GLM 5.2 vs GPT-6.1 Sol
- GLM 5.2 vs Claude Fable 5.1
- Claude Fable 5 vs GLM 5.2
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Claude Opus 5 vs GLM 5.2: questions
- Is Claude Opus 5 better than GLM 5.2?
- Claude Opus 5 (max) leads on quality: 65.5 vs 60.2. The BenchLeader Index combines every independent quality benchmark; Claude Opus 5 (max) is ahead overall as of 29 Oct 2026, but check the category scores for your use.
- Claude Opus 5 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 5 or GLM 5.2 better for coding?
- Claude Opus 5 scores higher in coding (70 vs 62 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1690 for Claude Opus 5 against 1606 for GLM 5.2.
- Which is faster, Claude Opus 5 or GLM 5.2?
- GLM 5.2 streams ~1.5× faster (94 vs 56 output tokens per second, Artificial Analysis 7-day median).
- Is Claude Opus 5 or GLM 5.2 better for agents and tool use?
- Claude Opus 5 scores higher in agentic tasks (64 vs 63 on the category index, where 50 is average). On τ³-Banking (AA), the top shared agents & tools benchmark, it is 42.1% for Claude Opus 5 against 34.6% for GLM 5.2.
More questions (4)
- Which is better for long documents?
- They are level on long-context benchmarks (65 vs 64). Both accept 1,000,000 tokens of context.
- Which has lower latency?
- GLM 5.2 answers first: 24.1 s against 43.9 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 vs GLM 5.2?
- Across the 35 benchmarks both have, Claude Opus 5 leads 34, GLM 5.2 leads 1. Every overall index that rates both puts Claude Opus 5 ahead.
- When was Claude Opus 5 released?
- Claude Opus 5 was released on 24 Jul 2026.