Home · 7

Claude Code v2.1.285 Adds Provider Restrictions, Desktop Integration, and Subagent Reliability Fixes
v2.1.285 introduces provider restrictions, desktop CLI flags, and resolves subagent permission and remote session bugs.
#claude-code#ai-agents#ai-news+1
Claude Code v2.1.284 Updates Auto-Mode Prompts, Gateway Spend Tracking, and MCP Reconnection
Claude Code v2.1.284 updates auto-mode prompts, adds gateway spend tracking, streamlines MCP reconnection, and patches SDK stability issues.
#claude-code#anthropic#ai-agents+3
Holo4 Series Launches Multi-Interface Agentic Models with Open Trajectories
New Holo4 and Holotron4 Nano models deliver cross-platform agent control via GUI, code, MCP, and APIs at lower inference costs.
#open-source-llms#ai-agents#automation+2
LangChain 1.4.3 Patches Tool Calls, Adds Bedrock Mantle Support, and Fixes Cache Routing
LangChain 1.4.3 patches critical tool call failures in create_agent, adds Bedrock Mantle support, and fixes cache settings for fallback models.
#ai-agents#langchain#tool-update+2
OpenAI Python SDK v3.20.0 Adds Agents Configuration, WebSocket Snapshots, and Transport Fixes
OpenAI Python SDK v3.20.0 introduces Agents credential options, incremental WebSocket snapshots, and critical queue and TLS bug fixes.
#OpenAI#python-sdk#agent-frameworks+3
Liquid AI Releases LFM2.5-VL-DSpark for Faster Vision-Language Inference
Liquid AI released LFM2.5-VL-DSpark, a 280M draft model that accelerates LFM2.5-VL-3B decoding by up to 3.13x with minimal memory overhead.
#AI#Open Source#LLM+4
Ollama v0.34.4 Speeds Up Structured Outputs and Enhances Apple Silicon Performance
Ollama v0.34.4 brings single-pass structured outputs for thinking models, Qwen 3.8 and Gemma 4 upgrades on Apple Silicon, and core bug fixes.
#Ollama#Open Source LLMs#Apple Silicon+4
OpenAI Python SDK v3.18.0 Adds Identifiers for GPT-6 Sol and Luna
OpenAI updates its Python library to version 3.18.0, adding official model identifiers for GPT-6 Sol and GPT-6 Luna.
#OpenAI#Python#SDK+3
n8n Introduces Dedicated Agents with Native Workflow and MCP Tool Integration
n8n launches standalone Agents featuring built-in session memory, approval controls, MCP support, and deep integration with standard workflows.
#n8n#AI Agents#Workflow Automation+3
Claude Code v2.1.283 Introduces Strict Model Controls, Prompt Auditing, and Gateway Optimization
Claude Code v2.1.283 adds exact model matching, denied models, gateway hint headers, a prompt audit command, and resolves SDK session recovery, MCP lifecycle, a
#claude-code#ai-agents#tool-update+2

No table of contents on this page

Table of Contents

No table of contents on this page