Compare / head-to-head
Claude Opus 4.8vsGLM-5.2
Claude Opus 4.8 leads 7 of 8 shared benchmarks. GLM-5.2 is 5.2x cheaper per token. Both offer a 1M-token context window.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
7 – 1
Claude Opus 4.8 leads
Cheaper per token
GLM-5.2
5.2x cheaper, input + output
Larger context
Tie
both 1M tokens
Provider uptime (30d)
100% · —
Anthropic · Z.ai
Claude Opus 4.8
Anthropic · released 2026-05-28
textvision
Quality
Benchmark matrix
8 shared · 8 only Claude Opus 4.8 · 17 only GLM-5.2| Benchmark | Claude Opus 4.8 | GLM-5.2 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 8 | ||||
| GPQA Diamond | 93.6% ↗ | 91.9% ↗epoch | +1.7 pt | Claude Opus 4.8 |
| AIME 2026 | 100% ↗matharena ⚠ | 90% ↗matharena ⚠ | +10 pt | Claude Opus 4.8 |
| FrontierMath Tier 4 v2 (Epoch AI run) | 56.1% ↗epoch | 29.3% ↗epoch | +26.8 pt | Claude Opus 4.8 |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 80% ↗epoch | 59.2% ↗epoch | +20.8 pt | Claude Opus 4.8 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 98.3% ↗epoch | 86.4% ↗epoch | +11.9 pt | Claude Opus 4.8 |
| SimpleQA Verified | 53% ↗epoch | 34.2% ↗epoch | +18.8 pt | Claude Opus 4.8 |
| SWE-Bench Pro | 69.2% ↗ | 62.1% ↗ | +7.1 pt | Claude Opus 4.8 |
| Terminal-Bench 2.1 | 74.6% ↗ | 81% ↗ | -6.4 pt | GLM-5.2 |
| Only Claude Opus 4.8 reports · 8 | ||||
| SWE-bench Verified | 88.6% ↗ | not reported | — | — |
| Humanity's Last Exam (no tools) | 49.8% ↗ | not reported | — | — |
| LMArena Elo | 1452.7 ↗ | not reported | — | — |
| BrowseComp | 84.3% ↗ | not reported | — | — |
| Humanity's Last Exam (with tools) | 57.9% ↗ | not reported | — | — |
| OSWorld-Verified | 83.4% ↗ | not reported | — | — |
| SWE-bench Multilingual | 84.4% ↗ | not reported | — | — |
| SWE-bench Multimodal | 38.4% ↗ | not reported | — | — |
| Only GLM-5.2 reports · 17 | ||||
| Agents' Last Exam (ALE-CLI) | not reported | 23.8% ↗ | — | — |
| AutomationBench v1.0.6 | not reported | 26.2% ↗ | — | — |
| CyberGym | not reported | 77.2% ↗ | — | — |
| DeepSWE v1.1 | not reported | 46.2% ↗ | — | — |
| ExploitBench | not reported | 24.4% ↗ | — | — |
| ExploitGym 2h | not reported | 29 ↗ | — | — |
| ExploitGym 6h | not reported | 39 ↗ | — | — |
| FrontierSWE | not reported | 67.5% ↗ | — | — |
| GDPval-AA v2 | not reported | 1508 ↗ | — | — |
| HLE (with tools) | not reported | 54.7% ↗ | — | — |
| NL2Repo | not reported | 48.9% ↗ | — | — |
| PostTrainBench | not reported | 31.7% ↗ | — | — |
| ProgramBench (Almost Solved) | not reported | 9.5% ↗ | — | — |
| SWE-bench Verified (Epoch AI run) | not reported | 78.7% ↗epoch | — | — |
| SWE-Marathon v1.1 | not reported | 19.4% ↗ | — | — |
| Terminal-Bench 3.0 | not reported | 4.6% ↗ | — | — |
| Toolathlon Verified | not reported | 59.9% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Claude Opus 4.8 minus GLM-5.2 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | Claude Opus 4.8 | GLM-5.2 | Edge |
|---|---|---|---|
| Context window | 1M tokens | 1M tokens | Tie |
| Max output | 128K tokens | 131K tokens | GLM-5.2 |
| Input price / 1M | $5 ↗ | $1.4 ↗ | GLM-5.2 |
| Output price / 1M | $25 ↗ | $4.4 ↗ | GLM-5.2 |
| Cached input / 1M | $0.5 ↗ | $0.26 ↗ | GLM-5.2 |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $30 | $5.8 | GLM-5.2 |
| Modalities | text · vision | text | Claude Opus 4.8 |
| Released | 2026-05-28 | 2026-06-24 | — |
| Cited benchmark scores | 17 | 26 | — |
Reliability
Provider status
Last 60 days · refreshed daily on this page · live on /statusMore matchups
Claude Opus 4.8 vs …
Models sharing the most benchmarksMore matchups
GLM-5.2 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Claude Opus 4.8 and GLM-5.2 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.