Update: But not for logprobs. Paper published today by @maksym_andr reports logprobs returned from OAI API are non-deterministic even with the seed parameter. Not sure how to reconcile this with sampled completions described as deterministic.
CODE
-
Building Restaurant Finder App: React Frontend with Amplication Backend
By
–
I am finally finishing a Simple Restaurant Finder Full Stack Web Application. Tech Stack Frontend: React.js Backend: @amplication Irony is backend with APIs, Auth, and everything created in 2 mins and it’s taking a lifetime for me to create a nice frontend.
-
Free RAG Course from Gen AI 360 Foundational Model Certification
By
–
Jump on our free RAG course from the Gen AI 360 Foundational Model Certification (Built in collaboration with @activeloop
, @towards_AI
, and the Intel Disruptor Initiative): https://
learn.activeloop.ai/courses/rag -
LangChain vs LlamaIndex vs OpenAI GPTs Comparison Guide
By
–
New video! @langchain vs. @llama_index vs. @OpenAI GPTs: Which one should you use? https://
youtu.be/g84uWgVXVYg -
Most Used Evaluation Metrics for AI Model Assessment
By
–
Most used evaluation metrics We see that most people still care primarily about correctness (and not other guardrails like toxicity or prompt injection) https://
docs.smith.langchain.com/evaluation/eva
luator-implementations
… -
Advanced Retrieval Strategies for AI Systems
By
–
Top Advanced Retrieval Strategies We primarily see people rolling their own retrieval, speaking to need to customize retrieval for each use case. Of general strategies, we see: 1. Self Query https://
python.langchain.com/docs/modules/d
ata_connection/retrievers/self_query
…
2. Hybrid Search
3. Contextual Compression -
Testing LLM Applications: Metrics and Evaluation Methods
By
–
How are people testing? Testing LLM Apps is hard. How are people doing it? 1. We see that 83% of test runs have some form of feedback, suggesting that most people are finding some metrics to eval (rather than just eyeball) 2. We see an average of 2.3 feedback per run,
-
Most Popular Vectorstores and Vector Databases 2026
By
–
Most Used Vectorstores Dominated by local vectorstores:
1. @trychroma 2. FAISS
4. @qdrant_engine 5. DocArray Vector-native databases next most popular
3. @pinecone 6. @weaviate_io Other Databases
7. PGVector
8. @supabase 9. @neo4j 10. @Redisinc 11. Azure
12. @AstraDB -
LangChain Expression Language Rises as Easy Chain Composition Tool
By
–
LangChain Expression Language Usage of LangChain Expression Language has been on the rise, as its emerged as an easy way to really compose chains together https://
python.langchain.com/docs/expressio
n_language/?ref=blog.langchain.dev
… -
LangSmith: Debugging and Production Monitoring for AI Applications
By
–
All stats are taken from LangSmith, from 2023-07-02 to 2023-12-11 http://
smith.langchain.com We see people using it for everything from debugging, to production monitoring, to testing & evaluation