FalkorDB
@falkordb.bsky.social
FalkorDB is a high-performance, multi-tenant graph database designed to address the limitations of traditional vector/search databases in Large Language Model (LLM) applications.
Pinned
FalkorDB
@falkordb.bsky.social
· Apr 28
🎉HackerNoon readers have voted FalkorDB the Startup of the Year 2024 for Tel Aviv. Our open-source property-graph database outscored more than 14k ventures on community impact, technical progress, and real-world traction. Thanks for your vote of confidence & happy graphing 👩💻
Reposted by FalkorDB
New integration alert 📢 📢: G.V() is now officially compatible with @falkordb.bsky.social !
We're bringing in first-class Cypher querying support along with high performance, and fully interactive graph visualization. But that's not everything!
Find out more at gdotv.com/blog/gdotv-s...
We're bringing in first-class Cypher querying support along with high performance, and fully interactive graph visualization. But that's not everything!
Find out more at gdotv.com/blog/gdotv-s...
G.V() is now compatible with FalkorDB
Descended from RedisGraph technology, FalkorDB is a versatile graph database that excels in multiple use cases. Its powerful GenAI-focused approach is ideal for applications such as GraphRAG, agentic ...
gdotv.com
September 4, 2025 at 3:07 PM
New integration alert 📢 📢: G.V() is now officially compatible with @falkordb.bsky.social !
We're bringing in first-class Cypher querying support along with high performance, and fully interactive graph visualization. But that's not everything!
Find out more at gdotv.com/blog/gdotv-s...
We're bringing in first-class Cypher querying support along with high performance, and fully interactive graph visualization. But that's not everything!
Find out more at gdotv.com/blog/gdotv-s...
FalkorDB v4.10.0 boosts graph performance with string interning, array indexing, & memory introspection for low-latency, predictable memory use in multi-tenant/real-time systems. Ideal for teams needing efficient memory & performance. Check it: github.com/FalkorDB/fal...
June 12, 2025 at 2:15 PM
FalkorDB v4.10.0 boosts graph performance with string interning, array indexing, & memory introspection for low-latency, predictable memory use in multi-tenant/real-time systems. Ideal for teams needing efficient memory & performance. Check it: github.com/FalkorDB/fal...
WOW, pitching at #SouthSummit was incredible.
We focused on a growing pain point: LLM hallucinations.
Specifically, why GraphRAG architecture delivers what VectorRAG can’t — trust, context, and accuracy at enterprise scale.
#GraphRAG #GenAI #FalkorDB #LLM #SouthSummit2025
We focused on a growing pain point: LLM hallucinations.
Specifically, why GraphRAG architecture delivers what VectorRAG can’t — trust, context, and accuracy at enterprise scale.
#GraphRAG #GenAI #FalkorDB #LLM #SouthSummit2025
June 5, 2025 at 8:42 AM
WOW, pitching at #SouthSummit was incredible.
We focused on a growing pain point: LLM hallucinations.
Specifically, why GraphRAG architecture delivers what VectorRAG can’t — trust, context, and accuracy at enterprise scale.
#GraphRAG #GenAI #FalkorDB #LLM #SouthSummit2025
We focused on a growing pain point: LLM hallucinations.
Specifically, why GraphRAG architecture delivers what VectorRAG can’t — trust, context, and accuracy at enterprise scale.
#GraphRAG #GenAI #FalkorDB #LLM #SouthSummit2025
Reposted by FalkorDB
3- @falkordb.bsky.social for AI-centric, Redis-backed graphs
May 15, 2025 at 1:54 PM
3- @falkordb.bsky.social for AI-centric, Redis-backed graphs
Presenting at Amazon Web Services (AWS) #AWSSummitTLV next week!
Come say hey & learn why what you can do to mitigate LLM #hallucinations with GraphRAG.
Come say hey & learn why what you can do to mitigate LLM #hallucinations with GraphRAG.
May 21, 2025 at 1:37 PM
Presenting at Amazon Web Services (AWS) #AWSSummitTLV next week!
Come say hey & learn why what you can do to mitigate LLM #hallucinations with GraphRAG.
Come say hey & learn why what you can do to mitigate LLM #hallucinations with GraphRAG.
In 2023, Klarna's AI chatbot aimed to replace customer service reps. FalkorDB tracked this, noting its evolution. Klarna's shift urges devs to audit AI, test graph retrieval for accuracy & scale, and prioritize structured retrieval for reliable generative AI.
May 15, 2025 at 1:30 PM
In 2023, Klarna's AI chatbot aimed to replace customer service reps. FalkorDB tracked this, noting its evolution. Klarna's shift urges devs to audit AI, test graph retrieval for accuracy & scale, and prioritize structured retrieval for reliable generative AI.
United Airlines used AI to streamline decisions with a secure data platform and flight ops pipelines. 90% adoption by managers shows disciplined AI boosts productivity. A knowledge graph could further cut costs, scale ops, and tighten loops for fewer delays. KG build guide below
May 13, 2025 at 3:58 PM
United Airlines used AI to streamline decisions with a secure data platform and flight ops pipelines. 90% adoption by managers shows disciplined AI boosts productivity. A knowledge graph could further cut costs, scale ops, and tighten loops for fewer delays. KG build guide below
Speaking at #ODSCEast 2025, May 13-15 in Boston & virtual! Topics: Graph DB for #machinelearning & why GraphRAG beats vectorRAG in GenAI. Slides & takeaways shared post-event. Join us! Free passes: odsc.com/boston/virtu... #ODSC2025
May 12, 2025 at 2:39 PM
Speaking at #ODSCEast 2025, May 13-15 in Boston & virtual! Topics: Graph DB for #machinelearning & why GraphRAG beats vectorRAG in GenAI. Slides & takeaways shared post-event. Join us! Free passes: odsc.com/boston/virtu... #ODSC2025
Graph-backed text-to-SQL demo at #KGC2025 today! Text-to-SQL boosts productivity: business users query in natural language, engineers cut reporting backlog. Semantic traversal, not matching, for production-grade results. Catch it later!
May 9, 2025 at 2:45 PM
Graph-backed text-to-SQL demo at #KGC2025 today! Text-to-SQL boosts productivity: business users query in natural language, engineers cut reporting backlog. Semantic traversal, not matching, for production-grade results. Catch it later!
Building a knowledge graph? At AWS DevOps day, we discussed #normalization: removing redundancies. E.g., "works_in" & "employed_by" are the same. Eliminate one to streamline your graph. #graphrag
May 7, 2025 at 11:04 AM
Building a knowledge graph? At AWS DevOps day, we discussed #normalization: removing redundancies. E.g., "works_in" & "employed_by" are the same. Eliminate one to streamline your graph. #graphrag
Building a knowledge graph? Ask:
Need LLMs with more context, less hallucination?
Want to visualize risk or track assets in real-time?
Bridging structured + unstructured data?
GraphRAG solves this. Start here: github.com/FalkorDB/Gra...
#GraphRAG #KnowledgeGraph
Need LLMs with more context, less hallucination?
Want to visualize risk or track assets in real-time?
Bridging structured + unstructured data?
GraphRAG solves this. Start here: github.com/FalkorDB/Gra...
#GraphRAG #KnowledgeGraph
May 6, 2025 at 10:06 AM
Building a knowledge graph? Ask:
Need LLMs with more context, less hallucination?
Want to visualize risk or track assets in real-time?
Bridging structured + unstructured data?
GraphRAG solves this. Start here: github.com/FalkorDB/Gra...
#GraphRAG #KnowledgeGraph
Need LLMs with more context, less hallucination?
Want to visualize risk or track assets in real-time?
Bridging structured + unstructured data?
GraphRAG solves this. Start here: github.com/FalkorDB/Gra...
#GraphRAG #KnowledgeGraph
Bank of America hit 2B+ AI interactions with >90% accuracy using small models, not massive LLMs. Their approach: robust retrieval architecture. Vector search is easy to prototype but fails in production with latency and opaque reasoning. Is your retrieval built for just a demo?
April 30, 2025 at 9:30 AM
Bank of America hit 2B+ AI interactions with >90% accuracy using small models, not massive LLMs. Their approach: robust retrieval architecture. Vector search is easy to prototype but fails in production with latency and opaque reasoning. Is your retrieval built for just a demo?
🎉HackerNoon readers have voted FalkorDB the Startup of the Year 2024 for Tel Aviv. Our open-source property-graph database outscored more than 14k ventures on community impact, technical progress, and real-world traction. Thanks for your vote of confidence & happy graphing 👩💻
April 28, 2025 at 9:37 AM
🎉HackerNoon readers have voted FalkorDB the Startup of the Year 2024 for Tel Aviv. Our open-source property-graph database outscored more than 14k ventures on community impact, technical progress, and real-world traction. Thanks for your vote of confidence & happy graphing 👩💻
Yesterday's 'Easily Achieve X0,000s Ops/sec with Multigraph Topology' workshop was awesome.
We started with a standalone machine with 16 cores = 26k queries per second. We tripled and then doubled the number of cores, and achieved linear scalability.
www.youtube.com/watch?v=LbeA...
We started with a standalone machine with 16 cores = 26k queries per second. We tripled and then doubled the number of cores, and achieved linear scalability.
www.youtube.com/watch?v=LbeA...
April 23, 2025 at 1:52 PM
Yesterday's 'Easily Achieve X0,000s Ops/sec with Multigraph Topology' workshop was awesome.
We started with a standalone machine with 16 cores = 26k queries per second. We tripled and then doubled the number of cores, and achieved linear scalability.
www.youtube.com/watch?v=LbeA...
We started with a standalone machine with 16 cores = 26k queries per second. We tripled and then doubled the number of cores, and achieved linear scalability.
www.youtube.com/watch?v=LbeA...
Here are 4 #RAG interview questions that separate surface-level prompt hackers from true retrieval engineers. These are the questions that will become standard in hiring panels for GenAI infrastructure roles. Getting these right is what separates operators from architects.
www.falkordb.com/try-free
www.falkordb.com/try-free
April 22, 2025 at 11:48 AM
Here are 4 #RAG interview questions that separate surface-level prompt hackers from true retrieval engineers. These are the questions that will become standard in hiring panels for GenAI infrastructure roles. Getting these right is what separates operators from architects.
www.falkordb.com/try-free
www.falkordb.com/try-free
GenAI's only as good as its retrieval layer. Survey shows: 85% deploy LLMs, 71% see output risks, 99% say human oversight’s key. Retrieval fails without structured data. Use graph databases to model relationships, route clean results to LLMs for fast, explainable answers. github.com/FalkorDB/Gra...
April 16, 2025 at 11:19 AM
GenAI's only as good as its retrieval layer. Survey shows: 85% deploy LLMs, 71% see output risks, 99% say human oversight’s key. Retrieval fails without structured data. Use graph databases to model relationships, route clean results to LLMs for fast, explainable answers. github.com/FalkorDB/Gra...
Large context windows (up to 1M tokens) won’t eliminate RAG:
- High latency, costs, and memory constraints.
- Performance degrades with longer texts.
- Privacy risks with unrestricted data access.
GraphRAG offers accurate, efficient retrieval. How are you managing these trade-offs?
#rag #llm
- High latency, costs, and memory constraints.
- Performance degrades with longer texts.
- Privacy risks with unrestricted data access.
GraphRAG offers accurate, efficient retrieval. How are you managing these trade-offs?
#rag #llm
March 17, 2025 at 12:59 PM
Pure semantic search struggles with relational queries. Hybrid retrieval—vector embeddings plus graph queries—is the fix. FalkorDB + LangChain + LangGraph handles this efficiently. How's your RAG stack handling this?
March 12, 2025 at 12:01 PM
Pure semantic search struggles with relational queries. Hybrid retrieval—vector embeddings plus graph queries—is the fix. FalkorDB + LangChain + LangGraph handles this efficiently. How's your RAG stack handling this?
Generative AI demands real-time reasoning, but traditional databases struggle with dynamic data. At NVIDIA's AI conference, FalkorDB will show how knowledge graphs power LLM workflows (e.g., GraphRAG) & fraud detection.
Think graph-native storage is the future?
Think graph-native storage is the future?
March 6, 2025 at 9:51 AM
Generative AI demands real-time reasoning, but traditional databases struggle with dynamic data. At NVIDIA's AI conference, FalkorDB will show how knowledge graphs power LLM workflows (e.g., GraphRAG) & fraud detection.
Think graph-native storage is the future?
Think graph-native storage is the future?
LLMs + memory = game-changer. FalkorDB's graph DB now integrates with LangChain, enabling AI agents to retain context across interactions. Dive into the tech: optimized queries, scalable architecture, and simplified dev workflow. Curious?
Check out the guide: www.falkordb.com/blog/buildin...
Check out the guide: www.falkordb.com/blog/buildin...
March 3, 2025 at 4:22 PM
LLMs + memory = game-changer. FalkorDB's graph DB now integrates with LangChain, enabling AI agents to retain context across interactions. Dive into the tech: optimized queries, scalable architecture, and simplified dev workflow. Curious?
Check out the guide: www.falkordb.com/blog/buildin...
Check out the guide: www.falkordb.com/blog/buildin...
NoLiMA: GPT-4o's accuracy drops from 99.3% to 69.7% when context expands from <1K to 32K tokens. Scaling context windows alone can't overcome the inability to model latent relationships. Property graphs = missing layer (explicit relationship encoding, metadata-aware retrieval).
February 19, 2025 at 2:45 PM
NoLiMA: GPT-4o's accuracy drops from 99.3% to 69.7% when context expands from <1K to 32K tokens. Scaling context windows alone can't overcome the inability to model latent relationships. Property graphs = missing layer (explicit relationship encoding, metadata-aware retrieval).
February 17, 2025 at 2:13 PM
Reminder: Need help writing effective cypher queries? We're hosting a webinar designed for developers, data scientists, and software architects who are either working with graph databases or exploring their potential.
Tuesday Feb 18 9:30AM - 10:15AM PST ↗️ Sign up: lu.ma/b2npiu4r
Tuesday Feb 18 9:30AM - 10:15AM PST ↗️ Sign up: lu.ma/b2npiu4r
February 6, 2025 at 11:53 AM
Reminder: Need help writing effective cypher queries? We're hosting a webinar designed for developers, data scientists, and software architects who are either working with graph databases or exploring their potential.
Tuesday Feb 18 9:30AM - 10:15AM PST ↗️ Sign up: lu.ma/b2npiu4r
Tuesday Feb 18 9:30AM - 10:15AM PST ↗️ Sign up: lu.ma/b2npiu4r
How much of your enterprise data do you send to an LLM?
#RAG #LLM #EnterpriseData #KnowledgeGraph #GraphRAG
#RAG #LLM #EnterpriseData #KnowledgeGraph #GraphRAG
February 5, 2025 at 3:37 PM
How much of your enterprise data do you send to an LLM?
#RAG #LLM #EnterpriseData #KnowledgeGraph #GraphRAG
#RAG #LLM #EnterpriseData #KnowledgeGraph #GraphRAG