LiteLLM proxy now routes OpenAI-style WebRTC realtime via HTTP, enabling credential exchange for real-time communication. Here's what changes for your architecture.

LiteLLM adds persistent video character management. This reduces redundant generation work and unlocks scalable video production workflows for builders.

Voiceflow's documented Conversations API enables programmatic agent interaction. Builders can now create custom interfaces and integrate voice agents into existing workflows.

Eleven Labs moves v3 out of alpha with 70+ language support, dialogue mode, and audio tags. Here's what changed and why it matters for your voice integration strategy.

ElevenLabs launches native iOS and Android apps, bringing production-grade voice synthesis to mobile. What this means for your product strategy.

ElevenLabs renamed its Conversational AI platform to Agents, signaling a shift toward positioning voice and text agents as discrete, deployable products rather than experimental features.

OpenHands reaches 1.0 with a new SDK designed for agent integration. Builders should evaluate whether this architecture supports their automation workflows.

GitHub Copilot's new Autopilot mode operates independently within VS Code, handling planning, code edits, and terminal execution. Builders need to understand the control model and integration implications.

Mistral AI launches Forge, a no-code agent platform positioning open-weight models against OpenAI and Anthropic. What builders need to know about deployment speed and data sovereignty tradeoffs.

Continue is addressing a critical friction point in multi-agent workflows by enabling separate chat sessions for agent handoffs, preventing context bloat in long sequential operations.