# LimitWatch

> Free, dated, sourced changelog of pricing, quota, rate-limit and plan changes at AI companies. Every entry cites an official first-party page.

## Data
- [All entries (JSON)](https://ailimitwatch.com/data/entries.json): CC BY 4.0
- [Entry schema](https://ailimitwatch.com/data/schema.json)
- [Tracked sources](https://ailimitwatch.com/data/sources.json)
- [JSON Feed](https://ailimitwatch.com/feed.json)

## Pages
- [Methodology](https://ailimitwatch.com/methodology/)
- [Disputes & corrections](https://ailimitwatch.com/disputes/)
- Contact: support@testedmcp.com
- [RSS feed](https://ailimitwatch.com/feed.xml)

## Monthly digests
- [2026-10](https://ailimitwatch.com/digest/2026-10/)
- [2026-09](https://ailimitwatch.com/digest/2026-09/)
- [2026-08](https://ailimitwatch.com/digest/2026-08/)
- [2026-07](https://ailimitwatch.com/digest/2026-07/)
- [2026-06](https://ailimitwatch.com/digest/2026-06/)
- [2026-05](https://ailimitwatch.com/digest/2026-05/)
- [2026-04](https://ailimitwatch.com/digest/2026-04/)
- [2026-03](https://ailimitwatch.com/digest/2026-03/)
- [2026-02](https://ailimitwatch.com/digest/2026-02/)
- [2026-01](https://ailimitwatch.com/digest/2026-01/)

## Company feeds
- [OpenAI (ChatGPT) ChatGPT / OpenAI API](https://ailimitwatch.com/companies/openai/feed.xml)
- [Anthropic (Claude) Claude / Claude API](https://ailimitwatch.com/companies/anthropic/feed.xml)
- [Google (Gemini) Gemini API](https://ailimitwatch.com/companies/google-gemini/feed.xml)
- [xAI (Grok) Grok API](https://ailimitwatch.com/companies/xai/feed.xml)
- [Cursor (Anysphere) Cursor](https://ailimitwatch.com/companies/cursor/feed.xml)
- [GitHub Copilot GitHub Copilot](https://ailimitwatch.com/companies/github-copilot/feed.xml)
- [Perplexity Perplexity / Sonar API](https://ailimitwatch.com/companies/perplexity/feed.xml)
- [DeepSeek DeepSeek API](https://ailimitwatch.com/companies/deepseek/feed.xml)
- [Together AI Together API](https://ailimitwatch.com/companies/together/feed.xml)
- [Fireworks AI Fireworks API](https://ailimitwatch.com/companies/fireworks/feed.xml)
- [Windsurf Windsurf](https://ailimitwatch.com/companies/windsurf/feed.xml)
- [v0 (Vercel) v0](https://ailimitwatch.com/companies/v0/feed.xml)
- [Amazon Bedrock (AWS) Amazon Bedrock](https://ailimitwatch.com/companies/aws-bedrock/feed.xml)
- [Ollama Ollama Cloud](https://ailimitwatch.com/companies/ollama/feed.xml)
- [Hugging Face Hugging Face Inference Providers](https://ailimitwatch.com/companies/huggingface/feed.xml)
- [Vercel AI Gateway Vercel AI Gateway](https://ailimitwatch.com/companies/vercel-ai-gateway/feed.xml)

## Changelog and guides
- [Changelog: all changes, newest first](https://ailimitwatch.com/changelog/)
- [How AI billing works](https://ailimitwatch.com/guides/how-ai-billing-works/)
- [Claude limit changes: Claude Duet](https://ailimitwatch.com/guides/claude-limit-changes/)
- [How Claude's weekly usage limits work](https://ailimitwatch.com/guides/claude-weekly-usage-limits/)
- [User reports (moderated, unconfirmed until an official source confirms)](https://ailimitwatch.com/reports/)
- [Aggregators vs direct: model price comparison](https://ailimitwatch.com/compare/aggregators-vs-direct/)
- [Price drops & deals](https://ailimitwatch.com/deals/) ([RSS](https://ailimitwatch.com/deals/feed.xml), [JSON](https://ailimitwatch.com/data/deals.json))
- [AI model prices compared](https://ailimitwatch.com/prices/): per-1M-token input/output prices with sources and last-verified dates

## Latest
- [2026-10-20 OpenAI (ChatGPT): 12-hour prompt caching option added; 12-hour cache writes cost 2.5x the input token price.](https://ailimitwatch.com/entries/2026-10-20-openai-openai-api-prompt-caching/)
- [2026-10-18 Fireworks AI: Small rate-limit tier threshold raised from <600B to <800B total parameters; ceilings unchanged.](https://ailimitwatch.com/entries/2026-10-18-fireworks-fireworks-serverless-rate-limit-tiers/)
- [2026-10-15 Anthropic (Claude): Fast mode for Claude Opus 5.5 left research preview and received a default rate limit in Start, Build and Scale tiers.](https://ailimitwatch.com/entries/2026-10-15-anthropic-claude-api-claude-opus-5-5-fast-mode/)
- [2026-10-13 xAI (Grok): Per-second request limit formula loosened from RPM / 60 to RPM / 48 with a minimum of 2.](https://ailimitwatch.com/entries/2026-10-13-xai-xai-api-request-rate-limits/)
- [2026-10-12 Amazon Bedrock (AWS): Quotas page now states applied quotas may be lower than listed defaults based on account age, usage and payment history, and increases aren't automatic.](https://ailimitwatch.com/entries/2026-10-12-aws-bedrock-amazon-bedrock-account-quotas/)
- [2026-10-08 GitHub Copilot: Org owners can bill Copilot code reviews to the organization instead of consuming members' Copilot quotas.](https://ailimitwatch.com/entries/2026-10-08-github-copilot-copilot-code-review/)
- [2026-10-07 Hugging Face: Free users no longer receive included monthly Inference Providers credits.](https://ailimitwatch.com/entries/2026-10-07-huggingface-hf-inference-providers-free-monthly-cred/)
- [2026-10-07 Fireworks AI: Adaptive rate-limit ceilings for Small serverless models doubled.](https://ailimitwatch.com/entries/2026-10-07-fireworks-fireworks-serverless-small-tier/)
- [2026-10-07 Anthropic (Claude): Max and Team plans now include monthly API credits for the Claude Platform.](https://ailimitwatch.com/entries/2026-10-07-anthropic-claude-max-team-plans/)
- [2026-10-07 Anthropic (Claude): Prompt cache read price on Claude Sonnet 5.5 lowered.](https://ailimitwatch.com/entries/2026-10-07-anthropic-claude-api-claude-sonnet-5-5/)
- [2026-10-06 OpenAI (ChatGPT): API usage tiers simplified from five to three (Build, Launch, Grow).](https://ailimitwatch.com/entries/2026-10-06-openai-openai-api-usage-tiers/)
- [2026-10-02 Together AI: Dynamic serverless rate limits that scaled with recent traffic were retired.](https://ailimitwatch.com/entries/2026-10-02-together-together-serverless/)
- [2026-10-02 Fireworks AI: Adaptive rate-limit ceilings for Small serverless models increased.](https://ailimitwatch.com/entries/2026-10-02-fireworks-fireworks-serverless-small-tier/)
- [2026-10-01 Perplexity: Decisions API model replaced by pplx-decider-v1.1-27b with lower input price (changelog dated October 2026, day not stated).](https://ailimitwatch.com/entries/2026-10-01-perplexity-perplexity-decisions-api-pplx-decider-v1/)
- [2026-10-01 Perplexity: Agent API low, medium, high and xhigh presets switched to Fast Search; web_search price lowered (changelog dated October 2026, day not stated).](https://ailimitwatch.com/entries/2026-10-01-perplexity-perplexity-agent-api-low-xhigh-presets-w/)
- [2026-10-01 Fireworks AI: Serverless price for DeepSeek V4.1 Flash increased (uncached input / cached input / output per 1M).](https://ailimitwatch.com/entries/2026-10-01-fireworks-fireworks-serverless-deepseek-v4-1-flash/)
- [2026-09-29 Windsurf: Users can sign in with ChatGPT and bill GPT model usage to an eligible ChatGPT Plus or Pro plan.](https://ailimitwatch.com/entries/2026-09-29-windsurf-devin-desktop-windsurf/)
- [2026-09-29 Amazon Bedrock (AWS): Per-model tokens-per-day quota replaced by a cross-model, per-account, per-Region daily token quota.](https://ailimitwatch.com/entries/2026-09-29-aws-bedrock-amazon-bedrock-bedrock-runtime/)
- [2026-09-28 xAI (Grok): Voice models brought under the spend-based rate limit tiers.](https://ailimitwatch.com/entries/2026-09-28-xai-xai-api-voice-rate-limit-tiers/)
- [2026-09-24 Anthropic (Claude): Billing resumed for refusals before any output in the bio, frontier_llm and reasoning_extraction categories.](https://ailimitwatch.com/entries/2026-09-24-anthropic-claude-api-refusal-billing/)
- [2026-09-23 Ollama: Off-peak per-token pricing introduced for cloud models.](https://ailimitwatch.com/entries/2026-09-23-ollama-ollama-cloud-off-peak-pricing/)
- [2026-09-23 Cursor (Anysphere): Time-limited (10-day) included usage credits for the new Rollouts bot on Teams and Enterprise.](https://ailimitwatch.com/entries/2026-09-23-cursor-cursor-teams-enterprise-rollouts/)
- [2026-09-22 Together AI: Serverless prices lowered for Qwen3.7-Max and Qwen3.8-Flash.](https://ailimitwatch.com/entries/2026-09-22-together-together-serverless-qwen3-7-max-qwen3-8-/)
- [2026-09-22 Anthropic (Claude): Five-hour usage limits increased on Pro, Max, Team and seat-based Enterprise plans at the Claude Opus 5.5 launch; subscription users also got a saveable rate limit reset. Weekly limits are not mentioned in the announcement.](https://ailimitwatch.com/entries/2026-09-22-anthropic-claude-pro-max-team-enterprise-plans/)
- [2026-09-02 xAI (Grok): grok-imagine-image-quality retires 2026-11-02; requests will be served by grok-imagine-image-2.0 (quality=low) at a lower per-image price.](https://ailimitwatch.com/entries/2026-09-02-xai-xai-api-grok-imagine-image-quality/)
- [2026-09-01 Perplexity: Agent API fast preset switched to Fast Search; web_search price per 1,000 invocations lowered (changelog dated September 2026, day not stated).](https://ailimitwatch.com/entries/2026-09-01-perplexity-perplexity-agent-api-fast-preset-web-sea/)
- [2026-09-01 Fireworks AI: Serverless adaptive rate-limit ceilings now vary by model size tier.](https://ailimitwatch.com/entries/2026-09-01-fireworks-fireworks-serverless/)
- [2026-08-31 Ollama: Ollama Cloud plans moved to per-token pricing with monthly included usage credits, replacing 5-hour and weekly limits; Team plan launched.](https://ailimitwatch.com/entries/2026-08-31-ollama-ollama-cloud-pro-max-team-plans/)
- [2026-08-31 GitHub Copilot: Model access for users with seats in several organizations is now determined only by the billing organization.](https://ailimitwatch.com/entries/2026-08-31-github-copilot-copilot-multi-org-users/)
- [2026-08-28 GitHub Copilot: Credit-card/PayPal Copilot Business and Enterprise seats move to upfront per-seat charges; prices unchanged.](https://ailimitwatch.com/entries/2026-08-28-github-copilot-copilot-business-enterprise/)
- [2026-08-27 Perplexity: Agent/Orchestrator API price for google/gemini-3.7-flash increased.](https://ailimitwatch.com/entries/2026-08-27-perplexity-perplexity-agent-api-gemini-3-7-flash/)
- [2026-08-21 OpenAI (ChatGPT): GPT-5.6 Sol price reduced to $4 input / $20 output per 1M tokens (promotional, at least through 2026-11-21).](https://ailimitwatch.com/entries/2026-08-21-openai-openai-api-gpt-5-6-sol/)
- [2026-08-21 Hugging Face: Monthly credits included with PRO, Team and Enterprise became general-purpose compute credits usable beyond Inference Providers.](https://ailimitwatch.com/entries/2026-08-21-huggingface-hf-inference-providers-compute-credits/)
- [2026-08-10 Anthropic (Claude): Claude Sonnet 5 introductory price made standard; scheduled increase to $3/$15 on 2026-09-01 cancelled.](https://ailimitwatch.com/entries/2026-08-10-anthropic-claude-api-claude-sonnet-5/)
- [2026-07-30 OpenAI (ChatGPT): Fast mode replaced the Priority Processing offering; for GPT-5.6 Sol it is priced at twice standard.](https://ailimitwatch.com/entries/2026-07-30-openai-openai-api-service-tiers/)
- [2026-07-30 OpenAI (ChatGPT): GPT-5.6 Luna price cut 80% and GPT-5.6 Terra price cut 20%.](https://ailimitwatch.com/entries/2026-07-30-openai-openai-api-gpt-5-6-luna-terra/)
- [2026-07-29 Ollama: Team plan listed: per-seat pricing with five-seat minimum.](https://ailimitwatch.com/entries/2026-07-29-ollama-ollama-cloud-team-plan/)
- [2026-07-28 Cursor (Anysphere): New Cursor Start plan for developers in India at INR 649 per month, billed in INR via UPI or card.](https://ailimitwatch.com/entries/2026-07-28-cursor-cursor-start-india/)
- [2026-07-21 Vercel AI Gateway: AI Gateway added service tiers (default, priority, flex) for OpenAI and Gemini models, billed per tier actually used.](https://ailimitwatch.com/entries/2026-07-21-vercel-ai-gateway-vercel-ai-gateway-service-tiers/)
- [2026-07-06 xAI (Grok): Request limits switched from requests per minute to requests per second, derived as RPM / 60.](https://ailimitwatch.com/entries/2026-07-06-xai-xai-api-request-rate-limits/)
- [2026-07-01 Perplexity: Agent API prices for GPT-5.6 Luna and Terra lowered; GPT-5.6 Sol Fast mode added at 2x standard token pricing (changelog dated July 2026, day not stated).](https://ailimitwatch.com/entries/2026-07-01-perplexity-perplexity-agent-api-gpt-5-6-models/)
- [2026-06-30 Amazon Bedrock (AWS): AWS updated its RPM quota guidance: one endpoint has no per-model RPM quotas; on bedrock-runtime RPM quotas are model-specific and some Claude Opus models have none.](https://ailimitwatch.com/entries/2026-06-30-aws-bedrock-amazon-bedrock-rpm-quotas/)
- [2026-06-28 Vercel AI Gateway: Free tier ($5/month credit) limited to a subset of models with lower per-model rate limits than paid tier.](https://ailimitwatch.com/entries/2026-06-28-vercel-ai-gateway-vercel-ai-gateway-free-tier/)
- [2026-06-27 Google (Gemini): Spend-based rate limits per rolling 10-minute window added alongside RPM/TPM.](https://ailimitwatch.com/entries/2026-06-27-google-gemini-gemini-api-spend-based-rate-limits/)
- [2026-06-15 xAI (Grok): A priority service tier became available on text inference endpoints; priority rates are billed only when priority is applied.](https://ailimitwatch.com/entries/2026-06-15-xai-xai-api-priority-processing/)
- [2026-06-09 Together AI: DeepSeek-V4-Pro serverless price lowered; cached-input pricing added for GLM-5.1 and Qwen3.5-397B-A17B.](https://ailimitwatch.com/entries/2026-06-09-together-together-serverless-deepseek-v4-pro-cach/)
- [2026-06-02 DeepSeek: deepseek-v4-pro 75%-off promotional prices became the listed standard prices.](https://ailimitwatch.com/entries/2026-06-02-deepseek-deepseek-api-deepseek-v4-pro/)
- [2026-05-29 Vercel AI Gateway: Pricing page lists charges for opt-in capabilities: Custom Reporting, team-wide provider allowlist, team-wide ZDR.](https://ailimitwatch.com/entries/2026-05-29-vercel-ai-gateway-vercel-ai-gateway-add-on-capability-char/)
- [2026-05-29 Together AI: Serverless prices increased for Qwen3.5-9B, Meta-Llama-3-8B-Instruct-Lite and Llama-3.3-70B-Instruct-Turbo.](https://ailimitwatch.com/entries/2026-05-29-together-together-serverless-qwen3-5-9b-llama-3-8/)
- [2026-05-27 Amazon Bedrock (AWS): AWS documented per-model input and output token quotas for the bedrock-mantle endpoint and how to request increases.](https://ailimitwatch.com/entries/2026-05-27-aws-bedrock-amazon-bedrock-bedrock-mantle-quotas/)