Back to Arena

Diffbot

by Diffbot Inc.

System Card

OrganizationDiffbot Inc.
Released2011-01
Architectureknowledge-base / web-crawl knowledge graph API
DetailsDiffbot uses computer vision and NLP to automatically extract structured data from web pages at scale, building the world's largest factual knowledge graph of public web entities (2B+ entities, 10T+ facts covering corporations, people, articles, products). The Knowledge Graph API enables structured querying of web-derived facts for RAG grounding and entity resolution. 155 patents filed.
Parameters—
Domainknowledge-graphrag-retrieval
Open SourceNo
WebsiteVisit
web-extractionentity-graphcomputer-visionknowledge-APIfacts

Capability Profile

Benchmark Scores

6 of 14 benchmarks
Long-Context Retrieval
2/5
RULER
71.151p
NIAH
no data
LooGLE
no data
LongBench
603p
∞Bench
no data
Multi-Turn Recall
0/2
LoCoMo
no data
MemoryBank
no data
Cross-Session Memory
1/1
LongMemEval
71.328p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
7579p
HotpotQA
76.983p
Agent Task Memory
0/1
AgentBench-Mem
no data
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
1/1
RAGAS
7272p
Sources:Diffbot vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Diffbot vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)Diffbot vendor documentation; evaluated on RAGAS: Automated Evaluation of Retrieval-Augmented Generation (Exploding Gradients, 2309)Diffbot vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Diffbot vendor documentation; evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)Diffbot vendor documentation; evaluated on RULER: What's the Real Context Size of Your Long-Context Language Models (NVIDIA, 2404)