AI Agents
AI Agent technology and applications
Articles
Gradium Launches stt-translate and s2s-translate: Real-Time⭐6
Gradium launches stt-translate and s2s-translate, real-time speech models that outperform GPT and Gemini in accuracy and latency with live browser translation a...
Nous Research Adds /learn to Hermes Agent’s Skills System,⭐7
Nous Research adds /learn to Hermes Agent, auto-generating reusable skills from docs, code, or logs as slash commands—no manual SKILL.md writing needed.
Using Graphify and NetworkX to Map Python Codebase Structure⭐7
Learn to map Python codebases offline using Graphify and NetworkX: detect god nodes, communities, and dependencies for architecture insights and refactoring.
16 Best Generative AI Coding Tools in 2026 Compared: Features,⭐10
Compare 16 top generative AI coding tools in 2026—features, strengths, and best fits for developers and enterprises.
DFlash Speculative Decoding Drafts Whole Token Blocks in⭐8
DFlash speculative decoding drafts entire token blocks in parallel, achieving up to 15x higher throughput on NVIDIA Blackwell GPUs.
Mistral OCR 4 Delivers Citation-Ready Structured Output for RAG,⭐8
Mistral OCR 4 delivers citation-ready structured output with bounding boxes, confidence scores, and 170-language support for self-hosted RAG, agentic, and
Datalab Releases lift: A 9B Open-Weights Vision Model That⭐8
Datalab releases lift, a 9B open-weights vision model that extracts structured JSON from PDFs using schemas, achieving 90.2% field accuracy in benchmarks.
How to Use NVIDIA Canary-1B-v2 for ASR, Translation, and⭐9
Learn how to use NVIDIA Canary-1B-v2 for ASR, translation, and SRT subtitle export in Python with this step-by-step guide.
Prime Intellect Unveils prime-rl 0.6.0 for Training⭐7
Prime Intellect releases prime-rl 0.6.0, boosting training of trillion-parameter MoE models for agentic RL workloads with better parallelization and scalability...
GLM-5.2 OpenAI-Compatible API: A Hands-On Guide to Reasoning⭐8
A hands-on guide to GLM-5.2's OpenAI-compatible API, covering reasoning effort, function calling, and long-context retrieval.