Introducing GPT-Live
OpenAI released GPT-Live-1 and GPT-Live-1 mini — full-duplex voice models that can listen and speak simultaneously; GPT-Live-1 mini replaces Advanced Voice Mode for all ChatGPT users globally.
Categories: OpenAI, Microsoft, Google, Alibaba, Vercel, DeepSeek, Perplexity (AI features)
29 noviniek
Results
Štvrtok 9. júla 2026
Streda 8. júla 2026
OpenAI Releases gpt-realtime-2.1 and gpt-realtime-2.1-mini — 25% Lower Latency and Reasoning for Voice Agents
OpenAI released new Realtime API models on July 6—gpt-realtime-2.1 cuts p95 latency by 25% and the mini variant brings reasoning and tool use at 6× lower cost.
Utorok 7. júla 2026
DeepSeek V4 Official Release Coming Mid-July With New Peak/Off-Peak API Pricing
DeepSeek's official V4 release is coming mid-July with a 1-million-token context window and the platform's first-ever peak/off-peak API pricing — calls during Beijing business hours will cost 2x the off-peak rate.
Pondelok 6. júla 2026
ChatGPT Business: Admin Controls for Plugins, New Dictation Speech-to-Text, and Workspace Agents Free Period Extended
ChatGPT Business adds admin controls for plugins with centralized management, a new speech-to-text dictation model in 25+ languages, and extended the Workspace Agents free period to July 6 when credit-based pricing begins.
Perplexity: Hybrid Local-Server Inference Orchestrator Arrives on Perplexity Computer in July
Perplexity AI introduces a hybrid orchestrator that automatically routes AI tasks between local device and cloud frontier models based on data sensitivity and compute requirements — no manual configuration needed.
Štvrtok 2. júla 2026
ChatGPT Rolls Out Updated Model Picker for Business and Enterprise — Instant Mode and New Naming
OpenAI updated ChatGPT's model picker for Business and Enterprise, renaming thinking tiers to Medium/High/Extra High, adding an Instant mode, and embedding the picker directly in the message composer across web, iOS, and Android.
Utorok 30. júna 2026
Realtime voice, speech, and transcription now supported on AI Gateway
Vercel's AI Gateway adds beta support for realtime voice agents, text-to-speech, and speech-to-text, so developers can build conversational voice apps without juggling separate providers. Audio models route through AI SDK 7 with the same observability, BYOK and spend controls as text models, at no extra Gateway fee.
Pondelok 29. júna 2026
Query Web Analytics from the Vercel CLI
New `vercel metrics` command exposes Web Analytics directly via the CLI. Supports filters, dimensions, and is positioned for coding agents to answer questions about UTM campaign performance or mobile-vs-desktop conversion.
Vercel Pro Teams Can Now Run Up to 500 Concurrent Builds
Concurrent build capacity for Pro teams jumped from 12 to 500. Enabled by default on Pro and Enterprise plans, billed on actual build minutes consumed.
Deploy Node Servers with Zero Configuration
Vercel auto-detects server.ts in the root or src/ and deploys it as a Node.js application without configuration. Supports native Node HTTP, Express, Koa, and NestJS.
Perplexity launches hybrid local-server inference orchestrator on Windows
Perplexity AI launches on Windows (July 2026) a hybrid inference orchestrator that automatically routes AI tasks between local device and cloud frontier models without user involvement. The system decides where each part of a task should execute.
Nedeľa 28. júna 2026
Alibaba releases Qwen-AgentWorld — Language World Models for general agents
Qwen-AgentWorld is the first model trained to predict the next environment state (not the next action) across seven agent domains. Two MoE sizes, 256K context, Apache 2.0.
Sobota 27. júna 2026
Deep Agents and OpenCode are now available in the AI SDK Harness
Vercel added two new adapters to the AI SDK Harness — LangChain Deep Agents and OpenCode — letting you run either coding-agent runtime through one unified HarnessAgent API inside a Vercel Sandbox.
Making private MCP servers reachable without making them public
OpenAI detailed how it preserved private network boundaries while supporting MCP streaming, authentication and an inspectable client, so enterprises don't have to expose internal MCP servers to the public internet.
Piatok 26. júna 2026
Mastering Codex Remote for engineering
OpenAI publishes a guide positioning Codex Remote as an 'engineering control plane' — phone as a controller for Codex sessions running on Macs, Windows machines or devboxes. Hosts, worktrees, goals, side chats, queued prompts.
Štvrtok 25. júna 2026
Vercel ships AI SDK 7 for production-ready TypeScript agents
AI SDK 7 brings approvals, durable WorkflowAgent, telemetry, MCP Apps, sandboxing and harness integrations for Claude Code, Codex and Pi.
Vercel AI Gateway adds GLM 5.2 Fast via Wafer at 170+ tokens/s
Z.ai's GLM 5.2 Fast is now available via Wafer on Vercel AI Gateway, averaging 170+ tokens/s with 2x the throughput of other providers.
Utorok 23. júna 2026
Sakana Fugu Ultra now available on AI Gateway
Vercel AI Gateway added Sakana AI's Fugu Ultra, a multi-agent orchestration model that routes work across 1-3 frontier-model agents and merges results, benchmarking comparably to Claude Mythos Preview and Fable 5.
Pondelok 22. júna 2026
Perplexity Comet for iOS: phone actions, iPad sidebar, Finance Deep Dive tab
Comet for iOS shipped its eighth round of improvements: one-tap phone-number actions (Call, FaceTime, Message, save to Contacts), a refined iPad sidebar, and Finance Deep Dive opening as a real browser tab.
New usage analytics and updated spend controls for enterprises
OpenAI added credit usage analytics — broken down by user, product, and model — plus monthly spend limits configurable per workspace, group, or individual for ChatGPT Enterprise and Edu. The goal: give CFOs and IT clearer visibility into who actually burns the AI budget.
Microsoft Copilot Cowork hits general availability with metered Copilot Credits pricing
Copilot Cowork moves from Frontier preview to worldwide GA — Microsoft 365 Copilot's agentic layer with usage-based billing and model picks across Claude Opus 4.8 / Sonnet 4.6 / GPT-5.5.
Vercel Connect: Secure access to external services for your agents
Vercel launched Connect, a new agent-infrastructure primitive providing short-lived scoped OAuth tokens for Slack, GitHub, Linear, Notion, Salesforce, Figma, Snowflake, Discord, plus generic OAuth/API-key and MCP server connectors.
Nedeľa 21. júna 2026
OpenAI ships Record & Replay for Codex on macOS
Demo a task once on a Mac and Codex bundles it into a parameterized skill — no scripting, no RPA, no low-code, just a demo.
Vercel Launches Konsistent for Unified Code Standards Across AI Agents and Humans, Plus Security Dashboard and Dry-Run Deployments
Vercel shipped Konsistent on July 1 to enforce uniform code standards for both AI agents and human developers, alongside a Security Dashboard in private beta and dry-run deployments for the Vercel CLI.
Gemini Omni Flash: Text-to-720p Video Generation Now in Public Preview via Gemini API
Google released gemini-omni-flash-preview, a high-performance multimodal model for high-speed video generation and conversational video editing — generating 3–10 second videos at 720p from text prompts or animating still images.
Sobota 20. júna 2026
Introducing OpenAI for Healthcare and improved health intelligence in ChatGPT
OpenAI launched OpenAI for Healthcare — a product suite for healthcare organizations with HIPAA support — and improved health intelligence in ChatGPT, which now sees 230M+ weekly health-related queries.
Piatok 19. júna 2026
OpenAI frontier models and Codex are now available on AWS
OpenAI made GPT-5.5, GPT-5.4 and Codex available via Amazon Bedrock — ending Azure's hyperscaler exclusivity for OpenAI APIs.
OpenAI GPT-5.6: ChatGPT Pro splits into three tiers, personal finance, and ChatGPT for PowerPoint
Beyond the GPT-5.6 model launch, OpenAI released new ChatGPT features: personal finance for Plus users in the US, ChatGPT for PowerPoint in Business plans, and signals suggest Pro plan splitting into Sol/Terra/Luna configurations.
Introducing eve — open-source agent framework
Eve is a TypeScript-native open-source framework where every agent = a directory of files (instructions.md + tools/ + skills/). Built in: durable execution, per-agent sandbox, human approvals, evals, OpenTelemetry. Plus the Sandbox runtime now goes up to 24h.