Weekly AI Limit Report

1. Jun 12

   Claude ↓ 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. source

2. Jun 12

   DeepSeek ↑ DeepSeek V4 Flash input price $0.0983 → $0.0980 per 1M tokens (+0%) source

3. Jun 12

   DeepSeek ↑ DeepSeek V4 Flash output price $0.20 → $0.20 per 1M tokens (+0%) source

4. Jun 12

   GitHub Copilot ↓ 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. source

5. Jun 12

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

6. Jun 11

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

7. Jun 11

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

8. Jun 11

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

9. Jun 11

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

10. Jun 11

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

11. Jun 11

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

12. Jun 11

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

13. Jun 11

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

14. Jun 10

    Cursor ↑ Reduced cost for Bugbot usage source

15. Jun 10

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

16. Jun 10

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

17. Jun 10

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

18. Jun 10

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

19. Jun 10

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

20. Jun 10

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

21. Jun 10

    v0 → 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. source

22. Jun 9

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

23. Jun 9

    v0 → Adds new model access to AI Gateway. source

24. Jun 8

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

25. Jun 8

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

26. Jun 8

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

27. Jun 8

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

Copy-paste for X, LinkedIn or your newsletter:

What changed in AI tool limits & pricing — past 7 days:

↓ Claude — 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. (Jun 12)
↑ DeepSeek — DeepSeek V4 Flash input price $0.0983 → $0.0980 per 1M tokens (+0%) (Jun 12)
↑ DeepSeek — DeepSeek V4 Flash output price $0.20 → $0.20 per 1M tokens (+0%) (Jun 12)
↓ GitHub Copilot — 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. (Jun 12)
↓ v0 — Requests routed to Claude Fable 5 via AI Gateway must fall back to other models; apps pinned to Fable 5 break with no notice. (Jun 12)
→ Claude — Changes the reset logic for weekly usage limits from a rolling window based on session start to a fixed account-specific schedule. (Jun 11)

Dated + sourced tracker → https://ailimit.watch/changes/