Compare / head-to-head
Claude Opus 4.8vsQwen3.8 27B
Claude Opus 4.8 leads 4 of 4 shared benchmarks. Qwen3.8 27B is 14.2x cheaper per token. Claude Opus 4.8 has the larger context window (1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
4 – 0
Claude Opus 4.8 leads
Cheaper per token
Qwen3.8 27B
14.2x cheaper, input + output
Larger context
Claude Opus 4.8
1M tokens
Provider uptime (30d)
100% · —
Anthropic · Qwen
Claude Opus 4.8
Anthropic · released 2026-05-28
textvision
Qwen3.8 27B
Qwen · released 2026-08-14
textvision
Quality
Benchmark matrix
4 shared · 12 only Claude Opus 4.8 · 0 only Qwen3.8 27B| Benchmark | Claude Opus 4.8 | Qwen3.8 27B | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 4 | ||||
| GPQA Diamond | 93.6% ↗ | 89.2% ↗ | +4.4 pt | Claude Opus 4.8 |
| LMArena Elo | 1452.7 ↗ | 1439.3 ↗ | +13.4 | Claude Opus 4.8 |
| SWE-Bench Pro | 69.2% ↗ | 61.7% ↗ | +7.5 pt | Claude Opus 4.8 |
| Terminal-Bench 2.1 | 74.6% ↗ | 73% ↗ | +1.6 pt | Claude Opus 4.8 |
| Only Claude Opus 4.8 reports · 12 | ||||
| SWE-bench Verified | 88.6% ↗ | not reported | — | — |
| AIME 2026 | 100% ↗matharena ⚠ | not reported | — | — |
| Humanity's Last Exam (no tools) | 49.8% ↗ | not reported | — | — |
| BrowseComp | 84.3% ↗ | not reported | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | 56.1% ↗epoch | not reported | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 80% ↗epoch | not reported | — | — |
| Humanity's Last Exam (with tools) | 57.9% ↗ | not reported | — | — |
| OSWorld-Verified | 83.4% ↗ | not reported | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 98.3% ↗epoch | not reported | — | — |
| SimpleQA Verified | 53% ↗epoch | not reported | — | — |
| SWE-bench Multilingual | 84.4% ↗ | not reported | — | — |
| SWE-bench Multimodal | 38.4% ↗ | not reported | — | — |
| Only Qwen3.8 27B reports · 0 | ||||
| Qwen3.8 27B reports nothing Claude Opus 4.8 does not. | ||||
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 Qwen3.8 27B 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 | Qwen3.8 27B | Edge |
|---|---|---|---|
| Context window | 1M tokens | 262K tokens | Claude Opus 4.8 |
| Max output | 128K tokens | — | — |
| Input price / 1M | $5 ↗ | $0.424 ↗ | Qwen3.8 27B |
| Output price / 1M | $25 ↗ | $1.7 ↗ | Qwen3.8 27B |
| Cached input / 1M | $0.5 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $30 | $2.12 | Qwen3.8 27B |
| Modalities | text · vision | text · vision | Tie |
| Released | 2026-05-28 | 2026-08-14 | — |
| Cited benchmark scores | 17 | 4 | — |
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
Qwen3.8 27B 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 Qwen3.8 27B on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.