# Claude Opus 4.8 vs GLM-5

> Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 46.1 on the Noometry Index. GLM-5 costs 6.5× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-opus-4-8-vs-glm-5
- Last updated: 2026-10-11
- Shared benchmarks: 36

## Summary

- They share 36 benchmarks with published results for both. Claude Opus 4.8 scores higher in 9 categories and GLM-5 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 4.8 leads 64.7 to 27.6.
- The biggest single-benchmark swing is ARC-AGI-2: 72.1% for Claude Opus 4.8 and 4.9% for GLM-5.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $5 / $25 for Claude Opus 4.8.
- Claude Opus 4.8 accepts more context: 1M tokens versus 205K.
- GLM-5 has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 60.7 | 46.1 |
| Rank | 13 | 66 |
| Context | 1M | 205K |
| Input $/M | $5 | $1 |
| Output $/M | $25 | $3.20 |
| Weights | Proprietary | Open |

## Coding

- Claude Opus 4.8: 59.9 (#12)
- GLM-5: 49.0 (#52)

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| LMArena WebDev | 1556 | 1434 |
| WeirdML | 82.9% | 48.2% |
| LMArena Coding | 1490 | 1461 |
| ALE-Bench | 1,564 | 765.62 |
| SWE-bench Verified | — | 72.1% |
| DeepSWE | 59% | — |
| FrontierCode | 46.5% | — |
| SWE-bench Verified (bash only) | — | 72.8% |
| SWE-bench Multilingual | — | 69.7% |
| SciCode | 53.5% | — |
| GSO | 47.1% | — |

## Agentic & Tool Use

- Claude Opus 4.8: 47.6 (#11)
- GLM-5: 31.1 (#71)

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| τ²-bench Banking | 39.7% | 9.8% |
| Vending-Bench 2 | 5,787 | 4,432 |
| Terminal-Bench | — | 52.4% |
| APEX-Agents | 48.9% | — |
| OSWorld 2.0 | 20.6% | — |
| Remote Labor Index | 8.3% | — |
| τ²-bench Airline | — | 82.5% |
| τ²-bench Retail | — | 73.7% |
| τ²-bench Telecom | — | 86.8% |
| DeepResearch Bench | 50.2% | — |
| PostTrainBench | 33.8% | — |
| GBAEval | 70.9% | — |
| GDP.pdf | 24% | — |
| LMArena Search | 1204 | — |

## Reasoning

- Claude Opus 4.8: 64.7 (#16)
- GLM-5: 27.6 (#116)

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| ARC-AGI-2 | 72.1% | 4.9% |
| SimpleBench | 64.8% | 53.2% |
| Kagi LLM Benchmark | 88.8% | 75% |
| NYT Connections (extended) | 91.1% | 74.8% |
| ARC-AGI-1 | 92.5% | 44.7% |
| Chess Puzzles | 34% | 10% |
| LMArena Hard Prompts | 1482 | 1452 |
| Epoch Capabilities Index | 158.21 | 145.83 |
| ForecastBench | 59.9 | 61 |
| CritPt | 20.9% | — |
| EnigmaEval | 23.5% | — |
| EBR-Bench | 28.6% | — |
| Mystery Game Puzzles | 36% | — |
| DTBench | 94.9% | — |
| LMCA | 57.5% | — |
| Surface Evolver Bench | 87.5% | — |
| Bench to the Future 3 | 0.14 | — |

## Math

- Claude Opus 4.8: 78.4 (#13)
- GLM-5: 46.4 (#71)

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| MathArena Final-Answer Competitions | 91.8% | 65.7% |
| OTIS Mock AIME 2024-2025 | 98.3% | 80% |
| LMArena Math | 1487 | 1440 |
| FrontierMath (Feb 2025 set) | 47.2% | 16.4% |
| FrontierMath Tier 4 (v1) | 31.3% | 2.1% |
| FrontierMath (Tiers 1-3) | 80% | — |
| FrontierMath Tier 4 | 56.1% | — |
| ProofBench | 69% | — |

## Knowledge

- Claude Opus 4.8: 61.3 (#29)
- GLM-5: 52.3 (#64)

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| GPQA Diamond | 91% | 87.8% |
| LMArena Expert | 1502 | 1454 |
| SimpleQA Verified | 53% | — |
| Vectara Hallucination Rate | — | 10.1% |

## Multimodal

- Claude Opus 4.8: 42.9 (#26)
- GLM-5: —

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| LMArena Vision | 1294 | — |
| Blueprint-Bench 2 | 14.5% | — |
| Furniture Assembly | 42.5% | — |
| LMArena Document | 1475 | — |

## Multilingual

- Claude Opus 4.8: 55.2 (#33)
- GLM-5: 53.7 (#58)

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| LMArena Non-English | 1450 | 1430 |
| LMArena Chinese | 1507 | 1511 |
| LMArena French | 1481 | 1455 |
| LMArena German | 1472 | 1445 |
| LMArena Japanese | 1440 | 1416 |
| LMArena Korean | 1432 | 1423 |
| LMArena Russian | 1474 | 1436 |
| LMArena Spanish | 1466 | 1454 |

## Instruction Following

- Claude Opus 4.8: 77.4 (#24)
- GLM-5: 75.2 (#67)

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| LMArena Instruction Following | 1476 | 1428 |

## Long Context

- Claude Opus 4.8: 45.4 (#35)
- GLM-5: 44.7 (#60)

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| LMArena Longer Query | 1483 | 1446 |
| CL-bench | — | 18.7% |

## Writing & Preference

- Claude Opus 4.8: 72.0 (#16)
- GLM-5: 66.0 (#38)

| Benchmark | Claude Opus 4.8 | GLM-5 |
|---|---|---|
| LMArena Text | 1461 | 1446 |
| LMArena Creative Writing | 1454 | 1439 |
| EQ-Bench Creative Writing | 1840 | 1601 |
| LMArena Multi-Turn | 1476 | 1456 |
| EQ-Bench 4 | 1281 | — |

## FAQ

### Is Claude Opus 4.8 better than GLM-5?

Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 46.1 on the Noometry Index. GLM-5 costs 6.5× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.

### Which is cheaper, Claude Opus 4.8 or GLM-5?

GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; Claude Opus 4.8 lists at $5 and $25.

### Is Claude Opus 4.8 or GLM-5 better for coding?

Claude Opus 4.8 scores higher on coding benchmarks: 59.9 versus 49.0 in the Noometry coding category.

### Which has the bigger context window?

Claude Opus 4.8 does, with 1M tokens against 205K.

### How many benchmarks do Claude Opus 4.8 and GLM-5 share?

36 benchmarks have published results for both models. Claude Opus 4.8 has 65 scored results on Noometry and GLM-5 has 45.