Every figure below comes from the launch directory and updates from X every 24 hours.
What Anthropic launched
Claude Opus 5.5 is the first model in Anthropic's Claude 5.5 family, released on 22 September 2026. Anthropic says it performs at the level of Claude Fable 5.1 on most work and costs 40% less to run than Opus 5.
It is the company's first release since it called for pacing the frontier, and it was tested before release by external evaluators including Frontier Design and METR.
The pricing is the headline for developers. Input and output tokens are $4 and $20 per million, 20% less than Opus 5.
Cache reads, which Anthropic says make up most of the cost of agentic and coding work, are $0.20 per million tokens, 60% less than Opus 5. Output is more than 30% faster. Usage limits went up on Pro, Max, Team and Enterprise plans on the same day.
Anthropic reports the best scores of any model to date on its automated behavioural audit, and says Opus 5.5 is less likely than recent models to take hard-to-reverse actions and more resistant to prompt injection.
Because it matches Claude Mythos 5.1 in biology and cybersecurity, it ships with the same safeguards as Fable 5.1. Claude Sonnet 5.5 and Claude Haiku 5.5 follow in the coming weeks.
What the launch post did
The announcement went out from @AnthropicAI on 22 September. It stands at 28,038,654 views, ranked 20 of the 1978 launches in the directory, with 97K likes, 9K reposts and 13K bookmarks.
| Anthropic launch | Views on X |
|---|---|
| Claude Opus 5.5, 22 September | 28M |
| Claude Fable 5.1 and Mythos 5.1, 1 September | 21M |
Three weeks apart, two frontier releases from the same account, and the second one out-travelled the first. Against the account's 1.9M followers the post reached 15 times its own following.
That multiple is low next to the launches from small accounts in the directory, and it always is for a lab this size: an account with millions of followers gets most of its views from them, and a founder with 1,500 gets almost all of hers from strangers. The number that places it is the rank.
How the post was built
Two sentences and a video. The first names the model and the family it opens. The second carries the two claims a reader can check, the Fable 5.1 comparison and the 40% cost cut.
There is no adjective in it. Anthropic's 1 September post for Fable 5.1 used a superlative instead, "the world's most advanced models for coding and knowledge work", and did fewer views.
The rest of the launch happened off the post. The announcement page carries the benchmarks, the safety results and the pricing table, and the thread and the replies carry the developers' own tests. The post itself asks the reader to click on one number.
What a company without 1.8M followers can take from it
- One comparison and one number. "Performs at the level of Fable 5.1, costs 40% less than Opus 5." A reader who knows the category can place the product in a second. Most launch posts spend that second on a slogan.
- Cut the adjectives. The post with none of them beat the post with "most advanced" by two million views, from the same account, three weeks apart.
- Price is a launch. A 40% cut is a headline on its own, and Anthropic put it in the second sentence rather than the announcement page.
- Your multiple will be higher. Anthropic reached 15 times its following. TypeSafe AI reached more than 300 times its own from 1,500 followers the week before. Small accounts get more reach per follower from the ranker, and the first hour decides how much.
Questions people are asking
What is Claude Opus 5.5?
Claude Opus 5.5 is Anthropic's newest model and the first in its Claude 5.5 family. Anthropic says it performs at the level of Claude Fable 5.1 on most work, costs 40% less to run than Opus 5, and scores best of any model on the company's automated behavioural audit.
When was Claude 5.5 released?
Claude Opus 5.5 was released on 22 September 2026. Anthropic said Claude Sonnet 5.5 and Claude Haiku 5.5 would follow in the coming weeks.
How much does Claude Opus 5.5 cost?
$4 per million input tokens and $20 per million output tokens, 20% less than Opus 5. Cache reads are $0.20 per million tokens, 60% less than Opus 5. On typical workloads Anthropic puts the total at 40% less than Opus 5.
How does Claude Opus 5.5 compare to Opus 5?
Anthropic describes it as a major step up: the new leading model, more than 30% faster at generating output, 40% cheaper on typical workloads, and more resistant to prompt injection. In one of its tests Opus 5.5 succeeded 39 of 40 times at cutting load times across a web app, where Opus 5 made smaller improvements that also changed the app's behaviour.
How did the Claude 5.5 launch do on X?
The launch post from @AnthropicAI stands at 28,038,654 views and ranks 20 of the 1978 AI launches Doomers tracks. Anthropic's previous launch, Claude Fable 5.1 and Mythos 5.1 on 1 September, stands at 21,044,338.
Did Doomers run the Claude 5.5 launch?
No. Anthropic ran its own launch. Doomers measures every AI launch on X in its directory and writes breakdowns of the largest ones.
The launches Doomers ran are on the case studies page.