Cost mode:

Discover natural groupings without external schema, name and order them coherently.

4 capabilities in this category.

Task-by-task breakdown

Topic Cluster Naming

Assigns a concise descriptive name (3-7 words) and a brief description (≤500 chars) to a cluster of semantically similar claims that has already been synthesised. Specific over generic — no 'Market …

ModelQuality (% of best)ConfidenceOverpay
GPT-5.4 Mini 94%MEDIUMbest value
Tencent Hy392%RANKED1.1x
MiniMax M394%RANKED1.2x
DeepSeek V4 Flash93%HIGH1.6x
GPT-5.6 Luna96%HIGH1.8x
DeepSeek V4 Pro best100%HIGH2.1x
Qwen 3.5 Flash90%HIGH2.2x
Qwen 3.7 Plus96%RANKED2.6x
Qwen 3.6 Plus97%HIGH2.6x
Gemini 3.1 Pro Preview91%HIGH3x
Claude Haiku 4.596%HIGH3.1x
NVIDIA Nemotron-3 Ultra 550B91%MEDIUM4.2x
Qwen 3.6 Flash95%RANKED4.3x
GPT-5.6 Terra95%MEDIUM4.5x
Gemini 3.5 Flash93%RANKED4.8x
Kimi K2.6100%RANKED5.7x
Claude Sonnet 4.698%HIGH6.7x
Claude Sonnet 595%RANKED8.7x
Meta Muse Spark 1.196%MEDIUM9.1x
GPT-5.599%RANKED9.8x
Grok 4.599%RANKED10x
GPT-5.6 Sol96%HIGH12x

Task detail →

Topic-to-Section Assignment

Assigns each merged topic in a batch to 1-3 publication sections, ordered by descending relevance. Index 0 is the PRIMARY section (where the topic appears on the client landing page); secondaries …

ModelQuality (% of best)ConfidenceOverpay
DeepSeek V4 Flash 91%MEDIUMbest value
Qwen 3.5 Flash92%HIGH2.7x
Tencent Hy391%MEDIUM2.8x
Qwen 3.6 Plus92%MEDIUM6.8x
Gemini 3.1 Pro Preview90%MEDIUM8x
Claude Sonnet 4.690%HIGH13x
Kimi K2.691%MEDIUM14x
Gemini 3.5 Flash best100%MEDIUM18x
GPT-5.595%HIGH23x
Meta Muse Spark 1.193%MEDIUM36x
GPT-5.6 Sol96%MEDIUM46x

Task detail →

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.