Fronset Benchmark

Claude Haiku 5.5 at a glance

Good enough on 17/54 tasks at the 90% bar. Best value on 4 tasks. Best fit for: Infrastructure & Utility, Long-form Content Generation, Social & Promotional Content. Doesn't qualify on any: Financial

Provider
Anthropic
Model name
claude-haiku-5-5
Cost mode:
Qualifies on
17 / 54 tasks (at 90% bar)
Best value on
4 tasks

Cost vs quality across all tasks

0%25%50%75%100%678910Quality score (6–10)Cost-efficiency vs best value (1.0 = best value)Voice Profile Generation — quality 8.75, 1.7x the cost of the best-value good-enough optionPromotional Campaign Generation — quality 7.45, this model IS the best-value good-enough option★Promotional Message Generation — quality 7.76, 2.0x the cost of the best-value good-enough optionCapability Category Matching — quality 8.56, 2.4x the cost of the best-value good-enough optionResearch Query Validation — quality 8.43, 2.7x the cost of the best-value good-enough optionMetadata Paragraph Rewriting — quality 7.95, 1.6x the cost of the best-value good-enough optionBench Hypothesis Post Generation — quality 8.73, 1.9x the cost of the best-value good-enough optionContent To Image Prompt Generation — quality 8.52, 1.6x the cost of the best-value good-enough optionStructured Content Summarization — quality 8.16, 1.2x the cost of the best-value good-enough optionPersona Policy Eligibility Check — quality 8.67, this model IS the best-value good-enough option★Profile Pool Matching — quality 8.28, 1.9x the cost of the best-value good-enough optionEngagement Opportunity Triage — quality 8.52, 2.7x the cost of the best-value good-enough optionSocial Post Portfolio Selection — quality 8.64, 1.4x the cost of the best-value good-enough optionReference Preserving Analytical Writing — quality 8.37, this model IS the best-value good-enough option★Profiled Document Structure Extraction — quality 8.46, this model IS the best-value good-enough option★Visual Theme Configuration Generation — quality 8.45, 2.7x the cost of the best-value good-enough optionShort Post Batch Relevance Scoring — quality 6.77, 8.6x the cost of the best-value good-enough option

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

TaskCategoryQuality (% of best)ConfidenceOverpay
Profiled Document Structure Extraction ★Structured Data & Fact Extraction95%MEDIUMbest value
Persona Policy Eligibility Check ★Relevance, Classification & Matching94%HIGHbest value
Promotional Campaign Generation ★Social & Promotional Content93%MEDIUMbest value
Reference Preserving Analytical Writing ★Long-form Content Generation91%MEDIUMbest value
Structured Content SummarizationContent Summarization & Synthesis91%HIGH1.2x
Social Post Portfolio SelectionRelevance, Classification & Matching99%RANKED1.4x
Content To Image Prompt GenerationInfrastructure & Utility96%RANKED1.6x
Metadata Paragraph RewritingInfrastructure & Utility95%HIGH1.6x
Voice Profile GenerationLong-form Content Generation91%HIGH1.7x
Profile Pool MatchingRelevance, Classification & Matching95%MEDIUM1.9x
Bench Hypothesis Post GenerationSocial & Promotional Content99%HIGH1.9x
Promotional Message GenerationSocial & Promotional Content91%HIGH2x
Capability Category MatchingRelevance, Classification & Matching96%MEDIUM2.4x
Engagement Opportunity TriageRelevance, Classification & Matching96%MEDIUM2.7x
Visual Theme Configuration GenerationLong-form Content Generation94%RANKED2.7x
Research Query ValidationInfrastructure & Utility97%MEDIUM2.7x
Short Post Batch Relevance ScoringRelevance, Classification & Matching93%MEDIUM8.6x

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.