search
LLM providers
Trends
- 1Thomson Reuters unveils Thomson, its own legal AI modelโMeet Thomson: The LLM built by Thomson Reuters
Thomson Reuters has introduced Thomson, a large language model built in-house and aimed at its legal and professional customers. The company positions the model as tailored to legal work, building on its Westlaw and CoCounsel products. Attention is likely to focus on how it compares with general-purpose models from OpenAI and other major AI providers.
- 2Adaptive routing for LLM traffic borrows TCP congestion controlโRouting LLM traffic across inference providers with TCP-style congestion control
A new approach applies TCP-style congestion control to route requests across multiple LLM inference providers, adapting to each provider's load, latency and failures in real time. The idea is drawing attention among developers building multi-provider AI systems, who see congestion-control signalling as a practical way to keep inference reliable and cost-effective as demand for model serving grows.
Repos
- unreallabsai/unreal-agent Async-first agent harness
- deepfates/imp declarative self-improving language-model programs for Elixir ๐
- sebastiancarlos/shobr ๐งโ๐ป The stealthiest, UNIX-iest, Job Search Automator
- XEonAX/blender-copilot A Copilot-style AI chat panel that lives inside Blender. The agent loop runs in Blender's own Python process, execu