Plans API Choosing a model If you're protecting your own systems, start with Kairo Blue : it reviews code and configuration, hunts threats, triages alerts and writes the fixes, and it never writes working exploits. Choose Kairo Red when you need a penetration test of systems you own or are authorised to test.
% $ # > X X *[ 0 =] : 1 x $ / & = !> ] } #; & & \! /+ > * ] ~ {# : x$ ? \/ @ 1 ! * |~ + ] 0 !# {# ?!~ + X % = + : = # * [ 1 [ / ! < & X + !* = & x $| / @= ; < | = ! : %X = 1 X \ > x 1 0 [ 1 < ? x?~= / + ~ : x = ? ! ^ & \ { $< * *< : X\#= ^ # * 1 * | ; * ]& $}~ ] 0 % x/ ! 0 $ /} = ; @ x > 1 ? ? $@ 1 ^ !+> * >]^ |=;@ \ \ ! [ | # & [ ; /
Our defensive model for code review, threat hunting, triage and fixes.
Access Early access
Input price —
Output price — View model ~ ! ~ = X ;^% x ] > $ > / ? +@ ! | X ;<0 11 { { }+ & =% @ { ;; # < } [ x | ? \| ~ :1{& { < 1 > ~ 0 # ; % ! ;] { [ + 0 : x { } ?& &{ 0 x < * % | ; X ] 1 * \! >] + \ 1 ;={ ^ | 0 =% 1 / { x /: \/ * % $ ~ * * x | \ { ^ # > ] ?|{ x ~[ ~ ? # ] ;# ! 1 <x = = [ X= / x ~ & 0 / ;= % % < | } X 1 = ~; % > * X =% : # !\ [ x X { : &[ 0 ]; ! 1|
Our offensive model for authorised penetration testing, inside a verified scope.
Access Verified security teams
Input price —
Output price — View model ? 0 ! /% = @ x [@ = | 0 $ ! @ * > / < \> * % 0 x1 ~> = = ! % * ] + < & # $ ++ | + = } x +% 1 0 # 1} &{ # 1 |[ X # < 1 } & * ! + ]; \^ ? : ;] ;# : % | \ { / # ! ? | & < } \ | X ~ ? ^1 X @ % \* ; [# +/ ^ ^ X~| 0 * > ;& ~/~ $ ! ] @] > / > 1 | @ x & X = $ = + / ? {} > \ [ ? ^= & ~ ]: { < [^ ~ * * : &@ stealth
A new model in preview
Access Preview · stealth
Input price —
Output price — View model Life sciences Models for research in biology and medicine.
In research Not on the API yet. Prices and model IDs are announced at launch.
vita Reads the literature, writes analysis code and drafts the documents research runs on.
In research · Early access with research partners next
More models Models from other companies, through the same API.
Z.ai Models by Z.ai. Requests are sent to Z.ai, which may store them.
For your hardest tasks
glm-5.3OpenAI Models by OpenAI. Requests are sent to OpenAI, which may store them.
DeepSeek Models by DeepSeek. Requests are sent to DeepSeek, which may store them.
Fast model, for fast questions.
deepseek-v4.1-flashAnthropic Models by Anthropic. Requests are sent to Anthropic, which may store them.
For your hardest tasks
claude-opus-5.5Moonshot Models by Moonshot. Requests are sent to Moonshot, which may store them.
Google Models by Google. Requests are sent to Google, which may store them.
Mistral Models by Mistral. Requests are sent to Mistral, which may store them.
Xiaomi Models by Xiaomi. Requests are sent to Xiaomi, which may store them.
Kairo Red is only available to verified security teams, on the API too. Every Red request is reviewed by hand.