Fronset Benchmark

Tencent Hy3 at a glance

Good enough on 15/54 tasks at the Balanced step. Best value on 2 tasks. Best fit for: Topic Organization & Clustering. Doesn't qualify on any: Financial Analysis & Trading Decisions.

Provider
OpenRouter
Model name
tencent/hy3
What the steps mean
Cost mode:
Qualifies on
15 / 54 tasks (at the Balanced step)
Best value on
2 tasks

Cost vs quality across all tasks

0%25%50%75%100%78910Quality score (7–10)Cost-efficiency vs best value (1.0 = best value)Factual Claim Refinement — quality 8.27, 5.5x the cost of the best-value good-enough optionAtomic Fact Claim Extraction — quality 7.43, 7.4x the cost of the best-value good-enough optionTopic Sequence Optimization — quality 8.12, 3.1x the cost of the best-value good-enough optionBench Hypothesis Post Generation — quality 8.30, 8.8x the cost of the best-value good-enough optionEngagement Opportunity Triage — quality 8.57, 24.9x the cost of the best-value good-enough optionPublic Response Generation — quality 7.84, 9.0x the cost of the best-value good-enough optionTopic Cluster Labeling — quality 8.14, 33.0x the cost of the best-value good-enough optionSocial Post Portfolio Selection — quality 7.98, this model IS the best-value good-enough option★Topic Section Assignment — quality 7.85, this model IS the best-value good-enough option★Content To Image Prompt Generation — quality 8.17, 10.8x the cost of the best-value good-enough optionModel Specific Prompt Adaptation — quality 8.36, 4.7x the cost of the best-value good-enough optionVisual Theme Configuration Generation — quality 7.99, 1.8x the cost of the best-value good-enough optionStructured Content Summarization — quality 8.66, 5.7x the cost of the best-value good-enough optionLanguage Identification — quality 9.85, 13.3x the cost of the best-value good-enough optionBatch Text Translation — quality 8.36, 7.3x 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 Balanced step are plotted.

Per-task breakdown

TaskCategoryQuality (% of best)ConfidenceOverpay
Topic Section Assignment ★Topic Organization & Clustering96%MEDIUMbest value
Social Post Portfolio Selection ★Relevance, Classification & Matching91%HIGHbest value
Visual Theme Configuration GenerationLong-form Content Generation91%MEDIUM1.8x
Topic Sequence OptimizationTopic Organization & Clustering92%HIGH3.1x
Model Specific Prompt AdaptationInfrastructure & Utility94%RANKED4.7x
Factual Claim RefinementInfrastructure & Utility96%HIGH5.5x
Structured Content SummarizationContent Summarization & Synthesis97%RANKED5.7x
Batch Text TranslationInfrastructure & Utility91%MEDIUM7.3x
Atomic Fact Claim ExtractionStructured Data & Fact Extraction96%MEDIUM7.4x
Bench Hypothesis Post GenerationSocial & Promotional Content94%MEDIUM8.8x
Public Response GenerationSocial & Promotional Content92%HIGH9x
Content To Image Prompt GenerationInfrastructure & Utility92%RANKED11x
Language IdentificationRelevance, Classification & Matching99%RANKED13x
Engagement Opportunity TriageRelevance, Classification & Matching97%RANKED25x
Topic Cluster LabelingTopic Organization & Clustering92%RANKED33x

Overpay — how much more you pay by running this model instead of the best-value model that meets the quality step 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.