Claude Opus 4.8 (archived)
Good enough on 7/51 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Relevance, Classification & Matching.
Cost vs quality across all tasks, as of 22 August 2026
within ~1.3× of the best-value model · 1.3–2× · >2× · ★ this model is the best-value pick on that task. Top-right = best quadrant. Only tasks where this model qualifies at the 90% bar are plotted.
Per-task breakdown, as of 22 August 2026
| Task | Category | Quality (% of best) | Confidence | Overpay |
|---|---|---|---|---|
| Claim Extraction | Structured Data & Fact Extraction | 97% | MEDIUM | 18x |
| Engagement Reply Draft | Social & Promotional Content | 91% | MEDIUM | 18x |
| Publication Title Generation | Content Summarization & Synthesis | 99% | RANKED | 18x |
| Claim Refinement | Infrastructure & Utility | 96% | HIGH | 23x |
| Engagement Reply Review | Relevance, Classification & Matching | 95% | MEDIUM | 49x |
| Engagement Triage | Relevance, Classification & Matching | 95% | RANKED | 71x |
| Language Detection | Relevance, Classification & Matching | 99% | RANKED | 78x |
Overpay — how much more you pay by running this model instead of the best-value model that clears the quality bar on that task (marked ★). "16x" means you overpay 16× — the same output for 16× the best-value good-enough option; ★ means this model is that option (no overpayment). Confidence — how sure we are about the quality score (more judgments + more agreement = higher confidence): RANKED many independent judges scored this model's outputs and their agreement is very high (most confident) — HIGH many judges have scored it and they mostly agree (well-pinned) — MEDIUM enough judges have weighed in to publish, but they disagree more than we'd like (treat with a small grain of salt). LOW-confidence cells are hidden everywhere on the site. See the methodology for the exact thresholds.