Fronset Benchmark

GPT-5.5 (archived)

Provider
OpenAI
Model name
gpt-5.5
Last benchmarked
Qualified on
27 / 51 tasks (at the 90% bar)
Best value on
0 tasks

Good enough on 27/51 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Financial Analysis & Trading Decisions, Infrastructure & Utility, Long-form Content Generation.

Cost vs quality across all tasks, as of 22 August 2026

0%25%50%75%100%8910Quality score (8–10)Cost-efficiency vs best value (1.0 = best value)Engagement Reply Draft — quality 8.30, 19.7x the cost of the best-value good-enough optionAuthor Voice Generation — quality 9.23, 105.8x the cost of the best-value good-enough optionAuthor Living-Person Safety Check — quality 9.14, 61.2x the cost of the best-value good-enough optionExecutive Summary Generation — quality 8.34, 65.8x the cost of the best-value good-enough optionTopic-to-Section Assignment — quality 8.55, 30.7x the cost of the best-value good-enough optionActivity Feed Blurb Generation — quality 8.55, 25.7x the cost of the best-value good-enough optionResearch Query Generation — quality 8.56, 42.2x the cost of the best-value good-enough optionPrompt Adaptation — quality 8.32, 110.4x the cost of the best-value good-enough optionTrading Recommendation — quality 8.61, 37.7x the cost of the best-value good-enough optionSEC Filing Analysis — quality 9.01, 19.7x the cost of the best-value good-enough optionX Post Selection — quality 8.72, 35.0x the cost of the best-value good-enough optionEngagement Triage — quality 8.50, 68.6x the cost of the best-value good-enough optionTranslation — quality 8.67, 89.6x the cost of the best-value good-enough optionTopic Cluster Naming — quality 8.58, 32.4x the cost of the best-value good-enough optionImage Prompt Generation — quality 8.45, 64.3x the cost of the best-value good-enough optionOnboarding Chapter Prompt Adaptation — quality 9.05, 109.3x the cost of the best-value good-enough optionTopic Grouping and Client Matching — quality 8.18, 33.6x the cost of the best-value good-enough optionAuthor Matching — quality 8.42, 43.3x the cost of the best-value good-enough optionClaim-Referenced Analyst Writing — quality 8.79, 136.0x the cost of the best-value good-enough optionInvestment Panel Voting — quality 8.56, 30.8x the cost of the best-value good-enough optionStructured Output Extraction — quality 8.96, 221.0x the cost of the best-value good-enough optionTopic Report Relevance Scoring — quality 8.64, 13.4x the cost of the best-value good-enough optionSEC S-1 Chunk Analysis — quality 9.00, 87.2x the cost of the best-value good-enough optionSocial Post Promotion — quality 8.23, 11.3x the cost of the best-value good-enough optionResearch Query Validation — quality 8.64, 42.5x the cost of the best-value good-enough optionVetted News Site Selection — quality 8.41, 3.3x the cost of the best-value good-enough optionLanguage Detection — quality 9.94, 37.9x 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, as of 22 August 2026

TaskCategoryQuality (% of best)ConfidenceOverpay
Vetted News Site SelectionRelevance, Classification & Matching98%MEDIUM3.3x
Social Post PromotionSocial & Promotional Content93%HIGH11x
Topic Report Relevance ScoringRelevance, Classification & Matching94%HIGH13x
Engagement Reply DraftSocial & Promotional Content98%HIGH20x
SEC Filing Analysis bestFinancial Analysis & Trading Decisions100%RANKED20x
Activity Feed Blurb GenerationSocial & Promotional Content98%RANKED26x
Topic-to-Section AssignmentTopic Organization & Clustering97%HIGH31x
Investment Panel VotingFinancial Analysis & Trading Decisions97%HIGH31x
Topic Cluster NamingTopic Organization & Clustering96%RANKED32x
Topic Grouping and Client MatchingRelevance, Classification & Matching96%MEDIUM34x
X Post SelectionRelevance, Classification & Matching99%RANKED35x
Trading RecommendationFinancial Analysis & Trading Decisions97%RANKED38x
Language DetectionRelevance, Classification & Matching99%RANKED38x
Research Query Generation bestInfrastructure & Utility100%MEDIUM42x
Research Query Validation bestInfrastructure & Utility100%HIGH42x
Author MatchingRelevance, Classification & Matching99%HIGH43x
Author Living-Person Safety Check bestRelevance, Classification & Matching100%MEDIUM61x
Image Prompt GenerationInfrastructure & Utility90%RANKED64x
Executive Summary GenerationContent Summarization & Synthesis100%MEDIUM66x
Engagement TriageRelevance, Classification & Matching93%RANKED69x
SEC S-1 Chunk AnalysisFinancial Analysis & Trading Decisions98%RANKED87x
TranslationInfrastructure & Utility92%RANKED90x
Author Voice GenerationLong-form Content Generation95%MEDIUM106x
Onboarding Chapter Prompt Adaptation bestInfrastructure & Utility100%RANKED109x
Prompt AdaptationInfrastructure & Utility94%RANKED110x
Claim-Referenced Analyst WritingLong-form Content Generation95%HIGH136x
Structured Output ExtractionStructured Data & Fact Extraction92%MEDIUM221x

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.