Haiku 4.5 (archived)
Good enough on 0/36 tasks at the 95% bar. Best value on 0 tasks. Doesn't qualify on any: Social & Promotional Content, Structured Data & Fact Extraction, Infrastructure & Utility, Content Summarization & Synthesis, Financial Analysis & Trading Decisions.
Cost vs quality across all tasks, as of 22 May 2026
This model doesn't qualify on any task at the 95% bar.
Per-task breakdown, as of 22 May 2026
| Task | Category | Quality (% of best) | Confidence | Overpay |
|---|---|---|---|---|
| Author Voice Generation | Long-form Content Generation | 96% | RANKED | — |
| Language Detection | Relevance, Classification & Matching | 100% | RANKED | — |
| Topic Cluster Naming | Topic Organization & Clustering | 96% | HIGH | — |
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.