Following its SpaceX deal, Anthropic doubles Claude Code's five-hour rate limits for paid plans and removes the peak hours limit reduction for Pro and Max plans
We've agreed to a partnership with SpaceX that will substantially increase our compute capacity.
Anthropic
Context & Ripple Effects
Anthropic tied a substantial increase in compute capacity from its SpaceX partnership to an immediate expansion of Claude Code access. Related coverage says the arrangement provides Anthropic access to Colossus 1 capacity, with a later expansion to Colossus 2 also reported.
The change matters because it converts infrastructure capacity into a visible product-policy shift: paid Claude users face fewer usage constraints, including during periods that had previously carried reduced limits.
First-order effects
Claude Code users on paid plans receive twice the five-hour usage allowance, while Pro and Max users no longer see peak-hour reductions.
Anthropic can serve more sustained coding workloads through Claude immediately, making the SpaceX capacity partnership operationally relevant to its customer offering rather than only a backend procurement move.
Second-order effects
Higher available usage raises the value of Anthropic’s paid tiers for developers and teams whose work was interrupted by prior caps, increasing pressure on rival coding-assistant providers to compete on dependable throughput as well as model capability.
The move makes Anthropic more dependent on delivering the newly contracted compute capacity reliably: capacity interruptions or demand that outpaces supply would directly affect a product promise now made to paying users.
Third-order effects
If major model providers increasingly turn large, dedicated infrastructure contracts into higher end-user limits, access to compute becomes a more direct determinant of AI-product differentiation and subscription economics.
The reported expansion from Colossus 1 to Colossus 2 suggests a possible shift toward longer-term, concentrated compute partnerships; whether that improves resilience or deepens supplier dependence will depend on how diversified providers’ capacity sources become.
The trend: Frontier AI vendors are translating large-scale compute procurement into less restrictive, more reliable access for high-intensity paid workloads, especially coding.
Effective today, we are: 1) Doubling Claude Code's 5-hour rate limits for Pro, Max, and Team plans; 2) Removing the peak hours limit reduction on Claude Code for Pro and Max plans; and 3) Substantially raising our API rate limits for Opus models.
Our agreement with @SpaceX means we will use all the compute capacity at their Colossus 1 data center. This will give us over 300 megawatts of additional capacity to deploy within the month.
With enough chips from SpaceXai, crisps from UK silicon vendors, etc. Anthropic finally has enough compute to stop rugpulling its customers. This week, anyway.
Anthropic is getting 220,000 Nvidia GPUs from SpaceX and has committed to spending $200 Billion on Google TPUs. Claude Code's 5-hour rate limits are doubling. Dario is finally getting aggressive about chips from every possible source to fight OpenAI Codex's growth. [image]
Anthropic loosening rate limits across the board by renting the entirety of Musk's Colossus-1 data center. Not really sure what the business strategy is here giving away $8-$13.50 for every $1 of sub revenue. Very funny. https://x.com/...
Anthropic is riding a big wave of good will (arguable IMO how deserved it is) and has decided to gamble that good will by partnering with the Musk's polluting CSAM-generating company. [embedded post]
The fine-print-omission from Claude Code's usage increase appears to be that weekly limits are *not* doubled. The 5-hour rate limit shrinking was indeed an efficiency blocker that finally convinced me to cancel, but being only able to get 4 full sessions a week doesn't compell m…
So, the weekly rate limits remain the same? “First, we're doubling Claude Code's five-hour rate limits for Pro, Max, Team, and seat-based Enterprise plans.” [image]
I'm old enough to remember when I was mean about Anthropic and a bunch of people turned up in my mentions to tell me not to because they were the good AI company. — www.anthropic.com/news/higher- ...