Wolfram Tech Now Available as Foundation Tool for LLM Systems
Stephen Wolfram has announced that Wolfram technology is now available as a foundation tool for LLM systems, positioning Wolfram Language as a general tool that provides deep computation and precise knowledge to supplement LLM foundation models.
Key Details
The announcement builds on three years of development since the initial Wolfram plugin for ChatGPT was released in March 2023. According to Wolfram, LLMs don't do deep computation and aren't precise, creating the need for a foundation tool that provides these capabilities.
Wolfram Language represents a general tool with general access to precise computation and knowledge, designed from the beginning to be a powerful medium for doing computation and representing things computationally. The technology provides:
- Broad and general technology that fits the breadth of LLM foundation models
- Access to precise computation and knowledge that LLMs lack
- A unified hub for connecting to other systems and services
- A medium for AIs to "think" and "reason" computationally
The development timeline includes:
- January 9, 2023: Wolfram posted "Wolfram|Alpha as the Way to Bring Computational Knowledge Superpowers to ChatGPT"
- March 2023: First Wolfram plugin for ChatGPT released
- Between these dates: Wolfram wrote "What Is ChatGPT Doing ... and Why Does It Work?"
Wolfram notes that three years have passed since ChatGPT's emergence, and it's now clearer that most growth in LLMs' practical value will come from how they are harnessed and connected. The technology is now positioned as a foundation tool that can be connected to LLM systems using streamlined protocols.
📖 Read the full source: HN LLM Tools
👀 See Also
Atlas Inference Engine Goes Open Source: Pure Rust + CUDA, 100+ tok/s on DGX Spark
Atlas is now open source — a Rust + CUDA inference engine that achieves 130 tok/s peak on Qwen3.5-35B (NVFP4) on a single DGX Spark, with no Python runtime and <2 minute cold start.
Claude Code Adds Auto Mode for Permission Decisions
Claude Code now has an auto mode that lets Claude handle permission decisions instead of requiring manual approval for every file write and bash command. This mode includes safeguards that check each action before it runs, with a classifier reviewing tool calls for potentially destructive actions.
Telegram Bot for Claude Code CLI Control from Mobile
A developer built a Telegram bot that bridges to the Claude Code CLI, allowing control via mobile commands like /commit, /code_review, and /simplify. The bot auto-discovers custom skills, processes photos/documents/voice notes, and supports group chat sessions.
Hybrid search with RRF improves AI memory system over pure vector search
An open-source AI memory system using PostgreSQL with pgvector found pure vector search insufficient for exact matches, so it added full-text search and merged results using Reciprocal Rank Fusion (RRF) with k=60, plus query enrichment via tokenizer.