Anthropic

Anthropic-Reranker-2024

reranker · Release Apr 15, 2024 · License n/a

A sophisticated reranking model that improves the relevance and quality of search engine results based on user intent.

rerankingsearchNLP

Updated today

Modalities

What goes in and what comes out.

Inputs

text

Outputs

text

Capabilities

result ranking, intent understanding, context awareness, A/B testing

Benchmarks snapshot

Structured JSON for reproducible comparisons.

{
  "relevance_score": 0.93
}

Related on GenAIWiki

Same provider, tooling that cites the model, or prompts tuned for it.

Anthropic

Claude 3 Opus

Claude 3 Opus enhances AI's conversational abilities with a broader understanding of context and intent, featuring a context window of 16k tokens for improved engagement in dialogues.

Anthropic

Claude 3.5 Sonnet

Balanced capability model emphasizing steerability, long-context reasoning, and safer default behaviors for agentic workflows.