AI Tool Changes

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro input price $0.90 → $0.69 per 1M tokens (-24%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $1.80 → $1.38 per 1M tokens (-24%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0855 → $0.0808 per 1M tokens (-5%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.17 → $0.16 per 1M tokens (-6%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro input price $1.04 → $0.90 per 1M tokens (-14%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $2.08 → $1.80 per 1M tokens (-14%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0886 → $0.0855 per 1M tokens (-3%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.18 → $0.17 per 1M tokens (-3%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro input price $1.60 → $1.04 per 1M tokens (-35%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $3.20 → $2.08 per 1M tokens (-35%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0809 → $0.0886 per 1M tokens (+10%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.16 → $0.18 per 1M tokens (+10%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $1.04 → $1.60 per 1M tokens (+54%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro output price $2.08 → $3.20 per 1M tokens (+54%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0855 → $0.0809 per 1M tokens (-5%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.17 → $0.16 per 1M tokens (-5%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $0.42 → $1.04 per 1M tokens (+149%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro output price $0.83 → $2.08 per 1M tokens (+149%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0795 → $0.0855 per 1M tokens (+8%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.16 → $0.17 per 1M tokens (+8%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro input price $0.62 → $0.42 per 1M tokens (-33%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $1.24 → $0.83 per 1M tokens (-33%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0834 → $0.0795 per 1M tokens (-5%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.17 → $0.16 per 1M tokens (-5%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro input price $0.74 → $0.62 per 1M tokens (-16%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $1.48 → $1.24 per 1M tokens (-16%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0868 → $0.0834 per 1M tokens (-4%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.17 → $0.17 per 1M tokens (-4%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro input price $0.87 → $0.74 per 1M tokens (-15%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $1.74 → $1.48 per 1M tokens (-15%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0780 → $0.0868 per 1M tokens (+11%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.16 → $0.17 per 1M tokens (+11%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0886 → $0.0780 per 1M tokens (-12%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.18 → $0.16 per 1M tokens (-12%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $0.79 → $0.87 per 1M tokens (+10%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro output price $1.58 → $1.74 per 1M tokens (+10%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $0.53 → $0.79 per 1M tokens (+50%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro output price $1.05 → $1.58 per 1M tokens (+50%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0574 → $0.0886 per 1M tokens (+54%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.11 → $0.18 per 1M tokens (+54%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $0.40 → $0.53 per 1M tokens (+33%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro output price $0.79 → $1.05 per 1M tokens (+33%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0517 → $0.0574 per 1M tokens (+11%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.10 → $0.11 per 1M tokens (+11%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro input price $0.45 → $0.40 per 1M tokens (-11%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $0.90 → $0.79 per 1M tokens (-11%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0713 → $0.0517 per 1M tokens (-27%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.14 → $0.10 per 1M tokens (-28%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro input price $1.60 → $0.45 per 1M tokens (-72%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $3.20 → $0.90 per 1M tokens (-72%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0826 → $0.0713 per 1M tokens (-14%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.17 → $0.14 per 1M tokens (-14%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0886 → $0.0826 per 1M tokens (-7%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.18 → $0.17 per 1M tokens (-7%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $1.32 → $1.60 per 1M tokens (+21%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $3.96 → $3.20 per 1M tokens (-19%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0826 → $0.0886 per 1M tokens (+7%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.17 → $0.18 per 1M tokens (+7%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $0.66 → $1.32 per 1M tokens (+100%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro output price $1.98 → $3.96 per 1M tokens (+100%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0798 → $0.0826 per 1M tokens (+4%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.16 → $0.17 per 1M tokens (+4%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro input price $1.32 → $0.66 per 1M tokens (-50%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Pro output price $3.96 → $1.98 per 1M tokens (-50%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0826 → $0.0798 per 1M tokens (-3%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.17 → $0.16 per 1M tokens (-3%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $1.17 → $1.32 per 1M tokens (+13%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro output price $2.34 → $3.96 per 1M tokens (+70%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0615 → $0.0826 per 1M tokens (+34%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.12 → $0.17 per 1M tokens (+34%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0741 → $0.0615 per 1M tokens (-17%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.15 → $0.12 per 1M tokens (-17%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.14 → $0.0741 per 1M tokens (-47%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.28 → $0.15 per 1M tokens (-47%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $0.63 → $1.17 per 1M tokens (+85%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro output price $1.26 → $2.34 per 1M tokens (+85%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro input price $0.43 → $0.63 per 1M tokens (+45%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Pro output price $0.87 → $1.26 per 1M tokens (+45%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0882 → $0.14 per 1M tokens (+59%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.18 → $0.28 per 1M tokens (+59%)

v0 STABLE ↑ Adds new image generation model capability to Vercel AI Gateway.

Claude WATCH ↓ Anthropic announced tightened biology-domain safeguards for Fable 5; announcement headline only, no published detail on which requests are affected.

GitHub Copilot CAUTION ↓ Users must now manually request Copilot code reviews instead of having them automatically assigned by the Code Quality feature.

GitHub Copilot CAUTION → Gives enterprise owners allowedMcpServers/deniedMcpServers keys to centrally control which MCP servers Copilot clients may run; restriction depends on the org’s own policy.

Notion AI STABLE ↑ Reduces friction for providing context to agents by enabling direct sharing from the document menu.

v0 STABLE ↓ Enterprise teams must migrate from Custom SIEM Log Streaming to the new supported destinations.

ChatGPT STABLE ↑ Increases model availability for free tier users, removing previous access restrictions.

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.14 → $0.0882 per 1M tokens (-37%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.28 → $0.18 per 1M tokens (-37%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek replaced the peak/off-peak notice with a pre-announced across-the-board price rise, described as a significant increase; specific plan and date still to be announced.

GitHub Copilot CAUTION ↑ Adds a new model option for users, subject to usage-based billing.

GitHub Copilot CAUTION → Gives org admins a knob to cap open pull requests for users without write access; not a vendor-imposed limit.

v0 STABLE ↑ Increased capacity for concurrent sandboxes and vCPU usage for Pro and Enterprise teams.

v0 STABLE ↑ Adds new model capability to AI Gateway

v0 STABLE [Pricing] ↑ Temporary price reduction for specific provider usage on AI Gateway.

GitHub Copilot CAUTION ↓ Users can no longer create new GitHub Spark apps or onboard as new users.

Mistral Le Chat CAUTION ↑ Adds a new model, Shieldstral, to the available lineup.

Cursor STABLE ↑ Adds new integration capabilities for Google Workspace users.

ElevenLabs CAUTION ↓ Limits the number of files held in memory for widget file-input, requiring summarization for larger sets.

ElevenLabs CAUTION ↓ Removes existing audio filtering capabilities from ElevenAgents configuration, potentially breaking workflows relying on these filters.

v0 STABLE ↑ Adds support for Qwen 3.8 Max model via Vercel AI Gateway.

DeepSeek CAUTION [Pricing] ↓ Pre-announced 2x price multiplier during peak hours (9:00-12:00 and 14:00-18:00 Beijing Time); the notice states the effective date is subject to a further official announcement, so not yet in force.

Devin WATCH → Allows users to customize Slack integration scope, reducing potential over-permissioning.

Devin WATCH → Changes the execution environment for MCP tools, potentially affecting network access and security posture.

Devin WATCH → Provides admins with granular control over tool availability across different environments.

Devin WATCH → Account-owned IP allowlists now also apply to automation webhooks; whether an integration breaks depends on the account's own allowlist configuration.

Gemini STABLE ↑ Google AI Pro and Ultra became available in Hong Kong and Macau.

GitHub Copilot CAUTION ↓ Users can no longer access Gemini 2.5 Pro and Gemini 3 Flash models in any Copilot experience.

GitHub Copilot CAUTION ↓ Automated workflows using 2FA-bypass tokens for sensitive actions will now fail and require interactive 2FA.

GitHub Copilot CAUTION ↓ Users must migrate away from deprecated models by September 1st, 2026, to maintain functionality.

Grok WATCH ↑ Adds new video generation capabilities including text-to-video and image-to-video.

Notion AI STABLE ↑ Adds automation capabilities to Custom Agents via meeting note triggers, reducing manual workflow overhead.

v0 STABLE → User-set spend caps and budget alerts for AI Gateway; a cost-control tool for customers, not a vendor-imposed cap.

v0 STABLE ↑ Increases usage limits for Laguna S 2.1, allowing higher volume tasks.

Bolt.new STABLE ↑ Adds a one-click security agent that scans, fixes and hardens generated apps before publishing.

ChatGPT STABLE [Pricing] ↑ Lower pricing for GPT-5.6 models on Luna and Terra tiers reduces cost barriers for enterprise deployment.

Gemini STABLE → Adds new capability for Chrome web browsing within Gemini Spark.

Gemini STABLE → Introduces new model/feature for robotics applications.

GitHub Copilot CAUTION → Gives enterprise admins a policy to restrict remote control to managed devices; whether a user loses access depends on their own org configuration.

Tabnine CAUTION ↓ Removes unauthorized bypasses for governance and agent feature controls, enforcing stricter policy compliance.

Cursor STABLE ↑ Adds support for Cursor on iPad, expanding platform availability.

Gemini STABLE ↑ Adds voice-based natural language interaction capabilities to the Gemini macOS application.

GitHub Copilot CAUTION ↑ Adds new agentic capabilities for code review to Pro, Pro+, Business, and Enterprise users.

GitHub Copilot CAUTION ↑ Simplifies model access for admins by enabling new models by default.

Grok WATCH ↑ New speech-to-speech model available, and the grok-voice-latest alias silently repoints to it on August 5, 2026.

v0 STABLE ↑ Increased capacity for Edge Config usage across all plans.

v0 STABLE ↑ Allows teams to scale beyond previous custom environment limits.

Windsurf CAUTION ↓ Agents are disabled when working in Restricted Mode, preventing automated assistance in sensitive or untrusted environments.

Windsurf CAUTION ↓ Specific models are no longer accessible within the Cascade interface, forcing users to switch to Devin Local.

Cursor STABLE → New product announcement; the blog and changelog listings disclose no change to limits, pricing, or model access.

GitHub Copilot CAUTION ↑ Adds a new model option for agentic coding workflows.

Tabnine CAUTION ↑ Restores GPT model access for Agent requests on Enterprise SaaS deployments after a period of intermittent failures.

ElevenLabs CAUTION ↓ Breaking change for API clients requiring schema updates to include conversation_id in WebRTC token responses.

GitHub Copilot CAUTION → Adds an admin governance control; GitHub itself removed no access. Whether an individual loses access depends on their own org's policy choice.

Perplexity STABLE ↑ Adds enterprise role-based access controls, a custom API credential vault, Brain access for Max subscribers and Opus 5.

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0938 → $0.14 per 1M tokens (+49%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.19 → $0.28 per 1M tokens (+49%)

Tabnine CAUTION ↑ Reverses the 1B tokens/month cap for on-prem deployments, ~3 weeks after it was introduced.

Cursor STABLE → Free tier now advertises Composer access in place of limited Tab completions. The pricing page does not say whether free Tab completions changed, so this is recorded as a swap in what is advertised, not a confirmed gain or loss.

Character.AI STABLE → Adds world-building context for Characters; beta with staged early access.

Claude WATCH ↑ New flagship Opus model at identical per-token pricing; Opus 4.8 still listed, so no forced migration or cost increase yet.

GitHub Copilot CAUTION ↑ Adds a new model option for users, increasing capability.

Midjourney STABLE ↑ New model version released; default-model status not stated in the announcement.

Notion AI STABLE [Pricing] → Introduces visibility into usage metrics for Notion Workers, preparing users for future billing transitions post-beta.

v0 STABLE ↑ Increases the maximum execution time for workflow steps to 30 minutes, allowing for longer-running background tasks.

Claude WATCH ↑ Enterprise tiers get a 500k context window on the default model (Team remains 200k), allowing larger document processing and longer session history.

GitHub Copilot CAUTION ↑ Adds autonomous background agent capabilities for Linear issue management.

GitHub Copilot CAUTION → Requires migration to stateless MCP protocol for GitHub MCP Server integrations.

GitHub Copilot CAUTION ↑ Expands Copilot coding agent capabilities to mobile Actions troubleshooting.

GitHub Copilot CAUTION ↑ Introduces review controls for autonomous agent issue management.

Grok WATCH ↑ Adds granular control over audio processing, reducing potential for missed speech in noisy environments.

v0 STABLE → Adds a model to AI Gateway with a time-limited free window that ends mid-August 2026.

Bolt.new STABLE → Adds a new feature for reusable context and workflow management, reducing setup friction.

Cursor STABLE → Addition of a new routing feature, likely impacting model selection or request handling.

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0938 → $0.0980 per 1M tokens (+4%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.19 → $0.20 per 1M tokens (+4%)

Lovable WATCH → Default-model swap for any app that does not pin a model — output behaviour can shift with no action by the builder (announced in the changelog).

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0980 → $0.0938 per 1M tokens (-4%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.20 → $0.19 per 1M tokens (-4%)

Gemini STABLE → New model variants introduced, expanding available model options.

Lovable WATCH ↑ Adds external sharing of internally published apps, gated to Business and Enterprise plans.

v0 STABLE ↑ Expands model selection available through the AI Gateway.

Claude WATCH ↑ Adds access to a new model version for Pro users.

ElevenLabs CAUTION ↓ Shorter default fallback window for the backup LLM cascade in ConvAI agents; remains user-configurable within the 2–15s range.

GitHub Copilot CAUTION [Pricing] → Improved visibility and management of AI credit allocation for cost centers.

Windsurf CAUTION ↑ Allows users to manage network access policies directly within the chat interface, reducing the need to switch contexts to the web app.

Gemini STABLE → A newly listed Gemini feature (Gemini Spark) is available in the US only at launch; not a restriction of previously available access.

Bolt.new STABLE → Adds new capability for generating presentations and interactive experiences.

Claude WATCH ↑ Reduces the minimum team size requirement for Claude Team plans, allowing smaller groups to access team-specific features.

GitHub Copilot CAUTION ↑ Adds custom instructions and configuration capabilities to Copilot code review, reducing setup friction.

Grok WATCH ↑ Expands model access to users in the EU region.

GitHub Copilot CAUTION ↑ Adds support for Xcode 27 on GitHub-hosted runners, expanding build environment capabilities.

Grok WATCH → Adds new automation capabilities to the Grok platform.

Notion AI STABLE ↑ Adds new capabilities for agents to manage schedules and calendar events directly via chat.

Tabnine CAUTION ↓ Prevents users from bypassing disabled model settings via client-side manipulation, ensuring strict adherence to administrative model access policies.

Devin WATCH ↓ Changes the metric for session sizing to focus exclusively on ACU consumption, potentially increasing cost visibility or impact for high-message-volume users.

GitHub Copilot CAUTION ↑ Expands the scope of secret scanning coverage for users.

Grok WATCH ↑ Increases accessibility to Grok Build by open-sourcing the project.

Tabnine CAUTION → Optional admin-configurable access control (Azure DevOps only): admins can require an account connection so users only receive context from repositories they are authorized to access. Off by default; enabled per account.

Tabnine CAUTION ↑ Increased context window allows for larger codebases and longer chat sessions.

Claude WATCH → Adds new specialized feature set for K-12 teachers.

Claude WATCH → Formalizes legal and data processing policies for K-12 educational usage.

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0770 → $0.0900 per 1M tokens (+17%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.15 → $0.18 per 1M tokens (+17%)

GitHub Copilot CAUTION ↑ Increases flexibility for users to bring their own keys and models to JetBrains IDEs.

GitHub Copilot CAUTION ↑ Adds new security analysis capability to the Copilot app interface.

Lovable WATCH ↓ Users on downgraded plans will have the 'Edit with Lovable' badge automatically re-enabled upon their next publish.

ElevenLabs CAUTION → Optional admin-set control: workspace admins can cap monthly credit consumption for new invitees. User-side budgeting capability, not a vendor-imposed limit.

ElevenLabs CAUTION ↓ Removes the ability to manually optimize streaming latency for TTS agents.

Perplexity STABLE ↑ Faster model performance and new capabilities like website publishing and memory improve utility.

Cursor STABLE → Adds new functionality for managing chat context and history.

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0900 → $0.0770 per 1M tokens (-14%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.18 → $0.15 per 1M tokens (-14%)

Gemini STABLE ↑ Adds a new organizational feature for study workflows in the Gemini app.

GitHub Copilot CAUTION ↑ Improved visibility into budget consumption via API, reducing manual overhead for tracking user limits.

HeyGen STABLE ↑ New features added including live avatar APIs and cloud rendering capabilities.

Lovable WATCH → Newly-added data-portability capability with built-in caps (5 GB, one export/project/24h) — net-new feature, not a reduction of an existing limit.

Lovable WATCH → Optional Enterprise admin control (off by default) to limit source-code zip downloads to admins/owners — admin empowerment, not vendor-imposed tightening.

Lovable WATCH → Adds an optional shorter (8h) enforced-SSO session duration alongside existing options — admin-set choice, not a vendor-imposed tightening.

Lovable WATCH → Enterprise member profiles are now private by default, requiring admin action to make them public.

Lovable WATCH → Workspace invitations now expire after 30 days, requiring re-invitation for stale pending requests.

Windsurf CAUTION ↑ Users gain access to the latest model version, improving performance capabilities.

GitHub Copilot CAUTION ↑ Expands access to Kimi K2.7 model to Business and Enterprise tiers.

GitHub Copilot CAUTION → A preview billing app is being retired and its functionality consolidated into the main billing UI with greater visibility; no loss of capability.

GitHub Copilot CAUTION ↑ Adds a new model option for agentic workflows in JetBrains IDEs.

Grok WATCH ↑ New model availability expands capabilities for coding and agentic tasks.

Grok WATCH [Pricing] → Establishes pricing structure for the new Grok 4.5 model.

Mistral Le Chat CAUTION ↑ Adds a new model capability for embodied navigation.

Notion AI STABLE ↑ Expands access to Notion Agents to mobile devices, increasing utility and accessibility.

Tabnine CAUTION ↓ A monthly fair-use ceiling of 1B tokens is newly imposed where none existed; the ceiling is high, so it binds only extreme-volume agentic workflows.

Tabnine CAUTION → Adds an opt-in admin governance control; not vendor-imposed on users. Only constrains users whose own org chooses to enable it.

Tabnine CAUTION ↓ Subagent runs now hit a default execution cap and can no longer bypass org policy locally or via Plan Mode.

ElevenLabs CAUTION ↑ Increased capacity for SIP REFER User-to-User Information data.

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0890 → $0.0900 per 1M tokens (+1%)

Grok WATCH ↓ Priority processing is no longer available for image and video inference endpoints, restricting high-priority scheduling to text inference only.

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0980 → $0.0890 per 1M tokens (-9%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.20 → $0.18 per 1M tokens (-8%)

Notion AI STABLE ↑ Allows integration of external AI agents into Notion workflows.

Notion AI STABLE ↑ Provides more model choices for different agent tasks.

Notion AI STABLE ↑ Expands platform support for developers building Notion Workers.

v0 STABLE ↑ Restores access to a specific model on the AI Gateway, reducing model-specific constraints.

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0900 → $0.0980 per 1M tokens (+9%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.18 → $0.20 per 1M tokens (+9%)

GitHub Copilot CAUTION ↓ Full removal of the GitHub Models service.

GitHub Copilot CAUTION → Allows administrators to enforce hard spending caps on agentic sessions.

GitHub Copilot CAUTION ↑ Adds a new model option to the Copilot model picker.

GitHub Copilot CAUTION ↑ Enables multimodal input capabilities for Copilot chat.

Claude WATCH [Pricing] → New Sonnet 5 flagship at introductory $2/$10 per MTok — cheaper than Sonnet 4.6 until Aug 31, 2026, then reverts to the same $3/$15. Favorable near-term; no price increase.

Claude WATCH ↑ Restores access to Fable 5 model previously suspended due to export controls.

Cursor STABLE → Introduces support for MCPs and Organizations within Team Marketplaces.

Gemini STABLE ↑ Adds new model options for developers and users.

GitHub Copilot CAUTION ↑ Users gain access to a newer version of the Claude Opus model.

GitHub Copilot CAUTION ↑ Adds a new model option for developers within Copilot.

GitHub Copilot CAUTION → New opt-in admin capability: enterprise admins can set a per-user AI credit budget on a cost center. Cost-governance tooling, not a platform-wide vendor limit.

v0 STABLE [Pricing] → Vercel Agent pricing changed; the changelog headline discloses no direction or magnitude.

v0 STABLE ↑ New model access added to Vercel AI Gateway.

v0 STABLE ↑ New model access added to Vercel AI Gateway.

Cursor STABLE ↑ Adds mobile access to the Cursor development environment.

Gemini STABLE ↑ Adds new functionality for Google AI Pro and Ultra subscribers in Google Meet.

v0 STABLE ↑ Allows for larger deployment artifacts in serverless functions.

v0 STABLE ↑ Expands model selection for AI Gateway users.

v0 STABLE ↑ Adds new capabilities for real-time audio processing.

v0 STABLE ↓ Sandboxes will now be reclaimed based on inactivity, potentially impacting long-running development environments.

Tabnine CAUTION ↑ Resolves a 404 error preventing access to specific Azure OpenAI models in IDE Chat/Agent.

Mistral Le Chat CAUTION [Pricing] ↓ Base OCR rate doubled ($2 → $4 / 1000 pages); callers pinned to mistral-ocr-latest pay 2x as the alias moved to OCR 4.

Mistral Le Chat CAUTION [Pricing] ↓ mistral-small-latest pricing raised: input +50% ($0.1 → $0.15), output doubled ($0.3 → $0.6) per M tokens.

Claude WATCH → Adds a new collaborative feature for teams.

GitHub Copilot CAUTION ↑ Allows users to bypass standard model provider constraints by using their own API keys.

Mistral Le Chat CAUTION ↑ Adds a new document intelligence model to the available toolset.

Tabnine CAUTION ↑ Provides visibility into model reasoning for self-managed models, improving debugging and transparency.

GitHub Copilot CAUTION ↑ Adds Claude as an agent provider in JetBrains IDEs, expanding model choice for users.

Grok WATCH ↑ Enables long-running autonomous tasks in Grok Build.

v0 STABLE → Adds a new model option for users of the AI Gateway, increasing model choice.

Tabnine CAUTION ↓ Users with very large context inputs may experience truncation or reduced context availability in agent interactions.

GitHub Copilot CAUTION ↑ Removal of preview status indicates general availability and increased stability for CLI usage.

GitHub Copilot CAUTION → The API now provides visibility into AI credit consumption per user, which may assist in monitoring usage-based billing.

Cursor STABLE → Adds new capabilities for cloud-based agentic workflows.

Cursor STABLE → Adds new functionality to Cursor Automations, expanding user capabilities.

GitHub Copilot CAUTION ↑ Removal of a hard block on new user registrations, transitioning to a gradual rollout.

GitHub Copilot CAUTION ↓ Users relying on Opus 4.6 (fast) must migrate to alternative models before the deprecation date.

Lovable WATCH [Pricing] → Consolidates billing into a single credit system, replacing separate dollar-based balances for Cloud and AI services.

Lovable WATCH → Adds native text-to-speech and speech-to-text capabilities to the platform.

Lovable WATCH → Adds private package management for Enterprise users.

Lovable WATCH → Opt-in admin governance control (off by default) — lets Enterprise admins enforce role-based editing; not a vendor-imposed restriction.

v0 STABLE ↑ Removes restrictions on deployment frequency or volume via the CLI.

Devin WATCH ↑ Increased capacity for enterprise knowledge storage.

GitHub Copilot CAUTION ↑ Simplifies model selection for users across all plans.

GitHub Copilot CAUTION ↑ Provides a new dedicated desktop interface for Copilot.

Midjourney STABLE ↑ Adds new functionality for V8.1, expanding user capabilities.

Replit STABLE → Reduces friction by enabling direct workflow integration between Claude and Replit.

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash input price $0.0900 → $0.0980 per 1M tokens (+9%)

DeepSeek CAUTION [Pricing] ↓ DeepSeek V4 Flash output price $0.18 → $0.20 per 1M tokens (+9%)

Gemini STABLE ↑ Adds new text-to-video generation capabilities via Gemini Omni.

GitHub Copilot CAUTION ↓ Product being sunset: new customers blocked from onboarding; existing users keep access until full retirement.

Luma Dream Machine STABLE ↑ Adds new workflow automation capabilities for users.

v0 STABLE ↑ Max Sandbox execution duration raised to 24 hours.

v0 STABLE → Adds new model access to AI Gateway.

ElevenLabs CAUTION → New optional music model; music_v1 flow remains available.

ElevenLabs CAUTION → Dev-facing enum change: async option replaced by fixed time windows; existing configs need a one-time migration.

ElevenLabs CAUTION ↑ Batch agent-test capacity raised 1000 → 5000 attached tests.

GitHub Copilot CAUTION → Improved visibility into active user counts for enterprise reporting.

Grok WATCH ↑ Allows users to opt-in to higher scheduling priority for inference requests, potentially reducing latency at an additional cost.

v0 STABLE ↑ Increased capacity for Hobby users to organize data across more Blob stores.

v0 STABLE ↑ Increases the maximum execution time for Vercel Functions from 800 seconds to 30 minutes for Pro and Enterprise teams.

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0980 → $0.0900 per 1M tokens (-8%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.20 → $0.18 per 1M tokens (-8%)

Claude WATCH ↓ Anthropic's newest and most capable models became unavailable to every customer overnight by federal directive — users building on Fable 5 / Mythos 5 must migrate with no notice. First known instance of the US government forcing a deployed frontier model offline.

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash input price $0.0983 → $0.0980 per 1M tokens (+0%)

DeepSeek CAUTION [Pricing] ↑ DeepSeek V4 Flash output price $0.20 → $0.20 per 1M tokens (+0%)

GitHub Copilot CAUTION ↓ Users who adopted Claude Fable 5 in Copilot lost access with no notice and must switch models. Reversal of the model addition made days earlier.

GitHub Copilot CAUTION ↑ Removes character constraints on repository-level custom instructions for Copilot code review.

GitHub Copilot CAUTION → Adds opt-in admin control to exclude content from Copilot code review; no existing capability removed.

v0 STABLE ↓ Requests routed to Claude Fable 5 via AI Gateway must fall back to other models; apps pinned to Fable 5 break with no notice.

Claude WATCH → Changes the reset logic for weekly usage limits from a rolling window based on session start to a fixed account-specific schedule.

GitHub Copilot CAUTION ↑ Adds opt-in automation capability for reasoning-based tasks; no existing capability removed.

Midjourney STABLE → Users are transitioned to the V8.1 model as the new default standard.

Replit STABLE → Adds a security layer that blocks malicious packages during installation, potentially preventing some package installs.

Suno STABLE ↑ Adds a new advanced stem separation feature restricted to Premier subscribers.

Tabnine CAUTION → New admin-controlled capability (opt-in) to enforce seat limits per team; not a vendor-imposed reduction.

Tabnine CAUTION → UI relocation — Test/Review tabs replaced by /test and /review slash commands; capability retained, workflow changes.

Tabnine CAUTION ↓ Users relying on local indexing must now use the remote Context Engine, which may impact privacy or offline capabilities.

Cursor STABLE [Pricing] ↑ Reduced cost for Bugbot usage

Devin WATCH → Users must migrate from the unofficial Figma integration to the official Figma MCP server.

GitHub Copilot CAUTION ↑ New model (Claude Fable 5) added across all plans; no existing model removed.

GitHub Copilot CAUTION ↑ Enterprises can now organize billing and usage tracking across more business units.

Grok WATCH ↑ Adds new capabilities for file storage, public URL generation, and direct integration with Imagine endpoints.

Replit STABLE ↑ Reduces repetitive prompting and context-setting for AI agents by allowing persistent custom instructions and skills.

Replit STABLE ↑ Expands access to the Replit-Databricks integration from a limited release to a public preview.

v0 STABLE [Pricing] → Billing-timing change (mid-cycle partial invoices) for on-demand usage; not a price increase — partial + end-of-cycle invoices sum to the same total.

Claude WATCH [Pricing] → New premium Fable 5 tier at $10/$50 per 1M tokens — 2x Opus; Opus pricing unchanged

Grok WATCH ↑ Adds new model access for developers via API.

v0 STABLE → Adds new model access to AI Gateway.

ElevenLabs CAUTION ↓ Users must migrate to v2 models before July 9, 2026, or lose access to these specific models.

ElevenLabs CAUTION ↓ Users must migrate to scribe_v2 or scribe_v2_realtime before July 9, 2026.

ElevenLabs CAUTION → Newly added privacy feature is restricted to enterprise customers for text-to-dialogue endpoints.

GitHub Copilot CAUTION → Admins can IP-restrict EMU account access (web, Git, API, tokens)

GitHub Copilot CAUTION ↓ Users can no longer access GPT-5.2 or GPT-5.2-Codex models in Copilot Chat, inline edits, or agent modes; enterprise admins must enable replacement models via Copilot policies.

Ideogram STABLE ↑ Self-hosters can drop API dependency/billing lock-in entirely

Windsurf CAUTION → Product identity shifts from code-first IDE to agent hub

Cursor STABLE [Pricing] ↓ Teams re-priced into two pools

GitHub Copilot CAUTION [Pricing] ↓ Switched to metered AI Credits, no fallback

GitHub Copilot CAUTION ↓ Agentic usage bills jump 10-50x

Runway STABLE ↓ Users lose unlimited slow-lane drafting; every generation now costs credits (though Max gives ~4x more credits than Unlimited's 2,250)

Gemini STABLE ↑ Partial walk-back after complaints

Gemini STABLE ↓ Moved to opaque compute-based limits

Gemini STABLE [Pricing] → New $100 Ultra tier; top tier cheaper

HeyGen STABLE ↓ Power users on unlimited forced to credit-based plans (effectively per-minute billing); migration irreversible

Perplexity STABLE ↓ Promo-account queries throttled

Claude WATCH ↑ 5h limits doubled; peak throttling reversed

HeyGen STABLE ↑ Removes a monthly credit drain for dubbing users; more predictable spend

Notion AI STABLE [Pricing] ↓ Agentic workflows now carry unpredictable monthly cost scaling with run frequency + data volume

Grok WATCH → Grok 4.3 released

Midjourney STABLE ↑ V8.1 generally available; HD ~33% more GPU time than SD

DeepSeek CAUTION ↑ New V4 models; default context raised to 1M

DeepSeek CAUTION ↓ Legacy API model endpoints scheduled for removal

GitHub Copilot CAUTION ↓ Opus models pulled from Pro tier

Devin WATCH [Pricing] ↓ Team cost shifts from $500 flat to $80 min + $40/seat (varies by team size); ACU predictability removed for self-serve users

Devin WATCH ↓ Teams using Devin Review in CI now face metered charges; Deep Mode queries add to spend

ChatGPT STABLE [Pricing] → New $100 Pro mid-tier

Claude WATCH ↓ Peak-hour throttling, ~7% of users hit session limits

Windsurf CAUTION [Pricing] ↓ Credits replaced by quotas; Pro price up 33%

Midjourney STABLE ↓ New model alpha-gated; Relax (unmetered) unavailable, styling jobs cost 4x

Claude WATCH ↑ Memory feature reached Free

Notion AI STABLE ↓ Hard daily cap blocks batch image generation; gated to Business+

Replit STABLE [Pricing] ↑ Core price cut 20%; Teams migrated to Pro

Grok WATCH → Multi-agent Grok 4.20 beta available

Replit STABLE [Pricing] ↓ Pricing overhaul; credits now expire; Teams retired

ChatGPT STABLE ↓ Older models removed from picker

Perplexity STABLE → Deep Research model upgraded

ChatGPT STABLE ↓ Ads introduced on Free/Go

Perplexity STABLE ↓ Deep Research quota slashed ~500/day → 20/mo

Perplexity STABLE ↓ Pro Labs halved 50 → 25/mo

Perplexity STABLE ↓ File uploads capped (unlimited → 50/week)

Kling AI STABLE ↓ Pro/Premier subscribers locked out of the flagship model at launch — must upgrade to Ultra (~$180/mo) for day-one access

ElevenLabs CAUTION → v3 promoted to GA

HeyGen STABLE ↓ Developers can't test/onboard without upfront payment; risk shifted to the developer

Gemini STABLE [Pricing] ↑ Cheaper AI Plus tier reached the US

Lovable WATCH ↑ Removes mid-cycle hard stops, but top-up credits cost more than base-plan credits; Free is excluded

ChatGPT STABLE [Pricing] ↑ Cheap $8 Go tier went global

Suno STABLE ↓ Free users can no longer export audio for use off-platform

v0 STABLE [Pricing] ↓ New individual developers lose the $20 solo tier; the next paid step up is ~50% more

Grok WATCH ↓ Image-generation policy tightened after regulatory pressure

HeyGen STABLE [Pricing] ↓ New teams' base cost jumps from ~$78 to $189/mo minimum; grandfathering is fragile

Claude WATCH ↑ Temporary holiday doubling

GitHub Copilot CAUTION ↓ Default $0 spend cap removed

Gemini STABLE ↓ Free Pro-model limit obscured/cut due to demand

Suno STABLE ↓ Paid creators lose IP-ownership assurance for licensing to film/TV/sync

Grok WATCH ↑ Grok 4.1 available free to all tiers

Cursor STABLE ↓ Pro free-trial window halved (2 weeks → 1 week)

Tabnine CAUTION [Pricing] ↓ Agentic capabilities cost +51% plus unbounded variable LLM consumption via pass-through

Sora STABLE ↓ Free-tier creators effectively blocked from iterative daily use

Sora STABLE ↓ Heavy users face per-generation cost after the daily free quota

Character.AI STABLE ↓ Complete removal of the primary use-case for under-18 users; CEO acknowledged expected churn

Cursor STABLE → Composer first-party model launched

Windsurf CAUTION ↑ Faster, stronger first-party model available

DeepSeek CAUTION ↓ Prior model retired; must move to V3.2-Exp

Sora STABLE ↓ Realistic human-likeness generation blocked by default unless the person opts in

Perplexity STABLE ↑ Comet browser opened to everyone

Gemini STABLE ↑ Daily image-generation allowance raised 30 → 100

GitHub Copilot CAUTION → Usage analytics dashboard added

Claude WATCH → Sonnet 4.5 released

DeepSeek CAUTION [Pricing] ↑ API token cost cut ~50%

ElevenLabs CAUTION ↓ Two legacy TTS models removed; pinned integrations break

Cursor STABLE [Pricing] → New mid-tier Pro+ ($60/mo, 3x Pro usage)

Kling AI STABLE [Pricing] ↑ Same credit balance produces ~40% more 1080p clips vs 2.1

Luma Dream Machine STABLE [Pricing] ↓ Flagship-model clips ~4x pricier than Ray2; budget users pushed to lower-quality models

Replit STABLE [Pricing] ↓ Agent 3 autonomy causes surprise cost overruns

Cursor STABLE ↓ Auto mode no longer free-unlimited

Replit STABLE → Agent 3 released

ChatGPT STABLE ↑ Plus 3-hour cap doubled after backlash

Synthesia STABLE [Pricing] ↑ Annual subscribers benefit; monthly subscribers see no relief — pressure to commit annually

DeepSeek CAUTION ↑ Context window restored (64K → 128K)

Claude WATCH ↓ New weekly ceiling on top of 5h limits

Bolt.new STABLE ↑ Central multi-user billing and access for collaborative use

Synthesia STABLE → Nominal minutes unchanged but credits drain faster across multi-feature use; tracking more complex

ChatGPT STABLE ↓ New per-model 3-hour caps at GPT-5 launch

Perplexity STABLE [Pricing] ↑ New $200 Max tier with uncapped research

Windsurf CAUTION → Ownership/roadmap now under Cognition

Grok WATCH [Pricing] → New $300/mo Heavy tier

Grok WATCH ↓ System-prompt safety incident

Cursor STABLE ↑ Refunds for unexpected usage

Bolt.new STABLE ↑ Less month-end pressure to burn tokens on paid plans

Replit STABLE [Pricing] ↓ Flat checkpoint price → variable effort-based cost

DeepSeek CAUTION ↓ Regulatory pressure to pull app in Germany (availability risk)

GitHub Copilot CAUTION ↓ Premium-request caps + overage billing enforced

Midjourney STABLE [Pricing] ↓ Video jobs burn ~8x the fast-GPU time of images

Cursor STABLE [Pricing] → New $200 Ultra tier

Cursor STABLE [Pricing] ↓ Request quota replaced by usage-based credits

Midjourney STABLE ↓ Copyright suit; possible forced content/output restrictions

ElevenLabs CAUTION [Pricing] ↑ v3 launched with a temporary 80% discount (expired end of June)

Windsurf CAUTION → First-party SWE-1 models added

Notion AI STABLE [Pricing] ↓ Cheapest path to full Notion AI jumps from ~$18 (Plus+add-on) to $20/seat Business; Free/Plus permanently lose AI

v0 STABLE [Pricing] ↓ Iterative/heavy users report credits depleting far faster than the old flat message model; cost now scales with token usage

Lovable WATCH [Pricing] → Plan options compressed to credit-slider increments; mid-tier fixed-credit options removed

v0 STABLE ↓ Free users hit message walls faster; the change appeared with no changelog or announcement

Claude WATCH [Pricing] → New Max tiers added

Devin WATCH [Pricing] ↑ Entry barrier dropped to $20, but per-ACU price rose; ~9 ACUs ≈ 2.25h of work

Tabnine CAUTION ↓ All free-tier users lost access; solo devs/students have no free path — hard paywall at $39/seat

Ideogram STABLE ↑ Model upgrade available across all tiers; no credit-cost increase announced

Sora STABLE ↑ EU/UK paying subscribers finally gained access

Mistral Le Chat CAUTION [Pricing] → Free users get an explicit daily ceiling and lose Mistral Large / Flash Answers

Tabnine CAUTION ↓ Individual devs/small teams priced out — nothing between (gone) free and $39/seat

Character.AI STABLE ↓ Teens face constrained model behavior; response-editing removed platform-wide

Devin WATCH [Pricing] → High entry barrier — only a $500/mo team commitment; individuals locked out

Sora STABLE ↓ EU/UK Plus & Pro subscribers had zero access to a feature they paid for

Mistral Le Chat CAUTION ↓ Broad disruption — integrations on Mixtral 8x7B etc. forced to migrate within ~4 months

Luma Dream Machine STABLE [Pricing] ↓ Free video generation paywalled across 25M+ users; commercial use locked behind Plus minimum

Luma Dream Machine STABLE ↓ Any creator monetizing content must subscribe at Plus ($29.99) minimum

Luma Dream Machine STABLE ↓ Heavy users who don't exhaust credits lose them each month

Poe STABLE [Pricing] ↑ Lower entry price for light users; the cheap tier is daily-capped vs the monthly-pooled higher tiers

DeepSeek CAUTION ↓ Main chat model's context window halved (128K → 64K)

Suno STABLE ↓ Free users locked out of the latest model; upgrade-or-wait dynamic

Character.AI STABLE ↓ Under-18 users see more content blocks; sensitive-topic roleplay increasingly interrupted

Mistral Le Chat CAUTION [Pricing] ↑ Large across-the-board API cost cut + zero-cost prototyping tier

Ideogram STABLE ↑ First programmatic access (enables pipeline/agentic use); new model free across tiers

Leonardo.ai STABLE → Roadmap risk of shifting toward enterprise/Canva integration over individual creators

Kling AI STABLE [Pricing] → Established the credit-per-tier structure still used today; free tier capped at 66 credits/day

Leonardo.ai STABLE ↑ Addresses 'use it or lose it' waste; but rollover bank is forfeited if you cancel

Suno STABLE ↓ Legal uncertainty for commercial users; downstream platforms wary of Suno-generated content

Leonardo.ai STABLE ↑ Free tier gained a new proprietary high-quality model

Synthesia STABLE ↓ Starter developers must upgrade (+$60/mo) for any API; Creator rate limits undocumented

Tabnine CAUTION ↑ Pro users gain a faster frontier model; free users remain on the proprietary model only

Ideogram STABLE [Pricing] ↑ Dedicated high-capacity tier for power users with predictable fast-queue allocation

Poe STABLE [Pricing] ↓ Fixed point budgets drain faster on popular creator bots that layer a markup on top of model cost

Mistral Le Chat CAUTION ↓ API users with hard-coded model strings had to change code across integrations

Synthesia STABLE ↓ Existing Personal users locked out at renewal; forced to Starter without legacy avatars

Replit STABLE [Pricing] ↓ Monthly Hacker/Pro tiers folded into an annual Core membership

Character.AI STABLE [Pricing] → Wait-time & speed advantages monetized; feature stratification begins

Replit STABLE ↓ Hosted apps that exceed egress limits get throttled

Poe STABLE → First hard per-model monthly message caps; free users hit a daily wall on premium models

Notion AI STABLE [Pricing] ↑ Any-tier teams could add AI without upgrading base plan

Replit STABLE ↓ Developers depending on the code-execution API lost it entirely