As we've consolidated our docs and shipped 1.0 versions of LangChain and LangGraph, we also rebuilt Chat LangChain from the ground up — now cleaner, faster, and more accurate. Chat LangChain is a chatbot that answers questions about LangChain resources, including our docs and
LLMS
-

Gemini Deep Research adds integration with Gmail, Docs, Drive, and Chat
By
–
Gemini Deep Research can now connect to your Gmail, Docs, Drive and Chat. "Deep Research will be pulling in information directly from your Drive, including Docs, Slides, Sheets and PDFs" Testing time
-

Kimi-K2 Reasoning Model Announcement
By
–
Kimi-K2 Reasoning model is coming soon https://
x.com/scaling01/stat
/scaling01/status/1986071916541870399
… -
Databricks Agent Bricks Enables Intelligent Agents Across Leading Models
By
–
You can now build intelligent agents with every leading model on Databricks, powered by Agent Bricks.
— Databricks (@databricks) 5 novembre 2025
We know that flexibility fuels innovation, and that every model brings its unique strengths – whether that’s cost, performance or speed. @OpenAI GPT-5 and @Google's Gemini… pic.twitter.com/OXm5wHvoNdYou can now build intelligent agents with every leading model on Databricks, powered by Agent Bricks. We know that flexibility fuels innovation, and that every model brings its unique strengths – whether that’s cost, performance or speed. @OpenAI GPT-5 and @Google
's Gemini -
RouteLLM API: Unified Access to Multiple Language Models
By
–
RouteLLM API is your one API for all your LLMs! Automatically route to the best model for your prompt, or call a specific one by name. Open-source LLMs at unbeatable prices, with automatic failover for closed-source models. Requires a ChatLLM Teams subscription (uses your
-

Groq Powers Paytm’s Inference for 300M Users in India
By
–
Groq Paytm
Inference in India just got faster and more intelligent.
300M users. 20M merchants.
Fueled by Groq. -
Tokenizer Approaches Impact LLM Performance on HellaSwag Benchmarks
By
–
If you measure downstream performance on HellaSwag rather than speedrun-equivalent loss, then different tokenizer approaches come out on top… The first run I did was much better on common-sense downstream, trained in equivalent time or better.
-
Tokenizer Training and Data Filtering Compliance Standards
By
–
Well, the tokenizer I used was trained on large quantities of data — I filtered the tokens based on yet more data from FineWeb. Question is if that's acceptable according to your rules…
-

Gemini 3 Preview Spotted on Vertex AI Platform
By
–
BREAKING : gemini-3-pro-preview-11-2025 has been spotted on Vertex AI (Not available yet). What we know so far – A preview version of Gemini 3 will be coming in November with 1M context window support. https://
x.com/legit_api/stat
/legit_api/status/1986066955191665043
…
