Posted by naruse on 25 Dec 2025
We are pleased to announce the release of Ruby 4.0.0. Ruby 4.0 introduces “Ruby Box” and “ZJIT”, and adds many improvements.
Ruby Box
Ruby Box is a new (experimental) feature to provide separation about definitions. Ruby Box is enabled when an environment variable RUBY_BOX=1 is specified. The class is Ruby::Box.
Definitions loaded in a box are isolated in the box. Ruby Box can isolate/separate monkey patches, changes of global/class variables, class/module definitions, and loaded native/ruby libraries from other boxes.
Expected use cases are:
- Run test cases in box to protect other tests when the test case uses monkey patches to override something
- Run web app boxes in parallel to execute blue-green deployment on an app server …
Posted by naruse on 25 Dec 2025
We are pleased to announce the release of Ruby 4.0.0. Ruby 4.0 introduces “Ruby Box” and “ZJIT”, and adds many improvements.
Ruby Box
Ruby Box is a new (experimental) feature to provide separation about definitions. Ruby Box is enabled when an environment variable RUBY_BOX=1 is specified. The class is Ruby::Box.
Definitions loaded in a box are isolated in the box. Ruby Box can isolate/separate monkey patches, changes of global/class variables, class/module definitions, and loaded native/ruby libraries from other boxes.
Expected use cases are:
- Run test cases in box to protect other tests when the test case uses monkey patches to override something
- Run web app boxes in parallel to execute blue-green deployment on an app server in a Ruby process
- Run web app boxes in parallel to evaluate dependency updates for a certain period of time by checking response diff using Ruby code
- Used as the foundation (low-level) API to implement kind of “package” (high-level) API (it is not designed yet)
For the detail of “Ruby Box”, see doc/language/box.md. [Feature #21311] [Misc #21385]
ZJIT
ZJIT is a new just-in-time (JIT) compiler, which is developed as the next generation of YJIT. You need Rust 1.85.0 or newer to build Ruby with ZJIT support, and ZJIT is enabled when --zjit is specified.
We’re building a new compiler for Ruby because we want to both raise the performance ceiling (bigger compilation unit size and SSA IR) and encourage more outside contribution (by becoming a more traditional method compiler). See our blog post for more details.
ZJIT is faster than the interpreter, but not yet as fast as YJIT. We encourage you to experiment with ZJIT, but maybe hold off on deploying it in production for now. Stay tuned for Ruby 4.1 ZJIT.
Ractor Improvements
Ractor, Ruby’s parallel execution mechanism, has received several improvements. A new class, Ractor::Port, was introduced to address issues related to message sending and receiving (see our blog post). Additionally, Ractor.shareable_proc makes it easier to share Proc objects between Ractors.
On the performance side, many internal data structures have been improved to significantly reduce contention on a global lock, unlocking better parallelism. Ractors also now share less internal data, resulting in less CPU cache contention when running in parallel.
Ractor was first introduced in Ruby 3.0 as an experimental feature. We aim to remove its “experimental” status next year.
Language changes
*nil no longer calls nil.to_a, similar to how **nil does not call nil.to_hash. [Feature #21047]
Logical binary operators (||, &&, and and or) at the beginning of a line continue the previous line, like fluent dot. The following code examples are equal:
if condition1
&& condition2
...
end
Previously:
if condition1 && condition2
...
end
if condition1 &&
condition2
...
end
Core classes updates
Note: We’re only listing outstanding class updates.
Array
Array#rfindhas been added as a more efficient alternative toarray.reverse_each.find[Feature #21678]Array#findhas been added as a more efficient override ofEnumerable#find[Feature #21678]
Binding
Binding#local_variables does no longer include numbered parameters. Also, Binding#local_variable_get, Binding#local_variable_set, and Binding#local_variable_defined? reject to handle numbered parameters. [Bug #21049]
Binding#implicit_parameters, Binding#implicit_parameter_get, and Binding#implicit_parameter_defined? have been added to access numbered parameters and “it” parameter. [Bug #21049]
Enumerator
Enumerator.produce now accepts an optional size keyword argument to specify the size of the enumerator. It can be an integer, Float::INFINITY, a callable object (such as a lambda), or nil to indicate unknown size. When not specified, the size defaults to Float::INFINITY.
# Infinite enumerator
enum = Enumerator.produce(1, size: Float::INFINITY, &:succ)
enum.size # => Float::INFINITY
# Finite enumerator with known/computable size
abs_dir = File.expand_path("./baz") # => "/foo/bar/baz"
traverser = Enumerator.produce(abs_dir, size: -> { abs_dir.count("/") + 1 }) {
raise StopIteration if it == "/"
File.dirname(it)
}
traverser.size # => 4
[Feature #21701]
ErrorHighlight
When an ArgumentError is raised, it now displays code snippets for both the method call (caller) and the method definition (callee). [Feature #21543]
test.rb:1:in 'Object#add': wrong number of arguments (given 1, expected 2) (ArgumentError)
caller: test.rb:3
| add(1)
^^^
callee: test.rb:1
| def add(x, y) = x + y
^^^
from test.rb:3:in '<main>'
Fiber
- Introduce support for
Fiber#raise(cause:)argument similar toKernel#raise. [Feature #21360]
Fiber::Scheduler
Introduce Fiber::Scheduler#fiber_interrupt to interrupt a fiber with a given exception. The initial use case is to interrupt a fiber that is waiting on a blocking IO operation when the IO operation is closed. [Feature #21166]
Introduce Fiber::Scheduler#yield to allow the fiber scheduler to continue processing when signal exceptions are disabled. [Bug #21633]
Reintroduce the Fiber::Scheduler#io_close hook for asynchronous IO#close.
Invoke Fiber::Scheduler#io_write when flushing the IO write buffer. [Bug #21789]
File
File::Stat#birthtimeis now available on Linux via the statx system call when supported by the kernel and filesystem. [Feature #21205]
IO
IO.select accepts Float::INFINITY as a timeout argument. [Feature #20610]
A deprecated behavior, process creation by IO class methods with a leading |, was removed. [Feature #19630]
Kernel
Kernel#inspect now checks for the existence of a #instance_variables_to_inspect method, allowing control over which instance variables are displayed in the #inspect string:
class DatabaseConfig
def initialize(host, user, password)
@host = host
@user = user
@password = password
end
private def instance_variables_to_inspect = [:@host, :@user]
end
conf = DatabaseConfig.new("localhost", "root", "hunter2")
conf.inspect #=> #<DatabaseConfig:0x0000000104def350 @host="localhost", @user="root">
[Feature #21219]
A deprecated behavior, process creation by Kernel#open with a leading |, was removed. [Feature #19630]
Math
Math.log1pandMath.expm1are added. [Feature #21527]
Pathname
- Pathname has been promoted from a default gem to a core class of Ruby. [Feature #17473]
Proc
Proc#parametersnow shows anonymous optional parameters as[:opt]instead of[:opt, nil], making the output consistent with when the anonymous parameter is required. [Bug #20974]
Ractor
Ractor::Port class was added for a new synchronization mechanism to communicate between Ractors. [Feature #21262]
port1 = Ractor::Port.new
port2 = Ractor::Port.new
Ractor.new port1, port2 do |port1, port2|
port1 << 1
port2 << 11
port1 << 2
port2 << 12
end
2.times{ p port1.receive } #=> 1, 2
2.times{ p port2.receive } #=> 11, 12
Ractor::Port provides the following methods:
Ractor::Port#receiveRactor::Port#send(orRactor::Port#<<)Ractor::Port#closeRactor::Port#closed?
As a result, Ractor.yield and Ractor#take were removed.
Ractor#join and Ractor#value were added to wait for the termination of a Ractor. These are similar to Thread#join and Thread#value.
Ractor#monitor and Ractor#unmonitor were added as low-level interfaces used internally to implement Ractor#join.
Ractor.select now only accepts Ractors and Ports. If Ractors are given, it returns when a Ractor terminates.
Ractor#default_port was added. Each Ractor has a default port, which is used by Ractor.send, Ractor.receive.
Ractor#close_incoming and Ractor#close_outgoing were removed.
Ractor.shareable_proc and Ractor.shareable_lambda are introduced to make shareable Proc or lambda. [Feature #21550], [Feature #21557]
Range
Range#to_set now performs size checks to prevent issues with endless ranges. [Bug #21654]
Range#overlap? now correctly handles infinite (unbounded) ranges. [Bug #21185]
Range#max behavior on beginless integer ranges has been fixed. [Bug #21174] [Bug #21175]
Ruby
- A new toplevel module
Rubyhas been defined, which contains Ruby-related constants. This module was reserved in Ruby 3.4 and is now officially defined. [Feature #20884]
Ruby::Box
- A new (experimental) feature to provide separation about definitions. For the detail of “Ruby Box”, see doc/language/box.md. [Feature #21311] [Misc #21385]
Set
Set is now a core class, instead of an autoloaded stdlib class. [Feature #21216]
Set#inspect now uses a simpler display, similar to literal arrays. (e.g., Set[1, 2, 3] instead of #<Set: {1, 2, 3}>). [Feature #21389]
Passing arguments to Set#to_set and Enumerable#to_set is now deprecated. [Feature #21390]
Socket
Socket.tcp&TCPSocket.newaccepts anopen_timeoutkeyword argument to specify the timeout for the initial connection. [Feature #21347]- When a user-specified timeout occurred in
TCPSocket.new, eitherErrno::ETIMEDOUTorIO::TimeoutErrorcould previously be raised depending on the situation. This behavior has been unified so thatIO::TimeoutErroris now consistently raised. (Please note that, inSocket.tcp, there are still cases whereErrno::ETIMEDOUTmay be raised in similar situations, and that in both casesErrno::ETIMEDOUTmay be raised when the timeout occurs at the OS level.)
String
Update Unicode to Version 17.0.0 and Emoji Version 17.0. [Feature #19908][Feature #20724][Feature #21275] (also applies to Regexp)
String#strip, strip!, lstrip, lstrip!, rstrip, and rstrip! are extended to accept *selectors arguments. [Feature #21552]
Thread
- Introduce support for
Thread#raise(cause:)argument similar toKernel#raise. [Feature #21360]
Stdlib updates
We only list stdlib changes that are notable feature changes.
Other changes are listed in the following sections. We also listed release history from the previous bundled version that is Ruby 3.4.0 if it has GitHub releases.
The following bundled gems are promoted from default gems.
-
ostruct 0.6.3
-
pstore 0.2.0
-
0.1.4 to v0.2.0
-
benchmark 0.5.0
-
logger 1.7.0
-
rdoc 7.0.2
-
6.14.0 to v6.14.1, v6.14.2, v6.15.0, v6.15.1, v6.16.0, v6.16.1, v6.17.0, v7.0.0, v7.0.1, v7.0.2, v7.0.3
-
win32ole 1.9.2
-
1.9.1 to v1.9.2
-
irb 1.16.0
-
reline 0.6.3
-
readline 0.0.4
-
fiddle 1.1.8
The following default gem is added.
- win32-registry 0.1.2
The following default gems are updated.
-
RubyGems 4.0.3
-
bundler 4.0.3
-
date 3.5.1
-
delegate 0.6.1
-
digest 3.2.1
-
3.2.0 to v3.2.1
-
english 0.8.1
-
0.8.0 to v0.8.1
-
erb 6.0.1
-
error_highlight 0.7.1
-
etc 1.4.6
-
fcntl 1.3.0
-
1.2.0 to v1.3.0
-
fileutils 1.8.0
-
1.7.3 to v1.8.0
-
forwardable 1.4.0
-
1.3.3 to v1.4.0
-
io-console 0.8.2
-
0.8.1 to v0.8.2
-
io-nonblock 0.3.2
-
io-wait 0.4.0
-
0.3.2 to v0.3.3, v0.3.5.test1, v0.3.5, v0.3.6, v0.4.0
-
ipaddr 1.2.8
-
json 2.18.0
-
2.9.1 to v2.10.0, v2.10.1, v2.10.2, v2.11.0, v2.11.1, v2.11.2, v2.11.3, v2.12.0, v2.12.1, v2.12.2, v2.13.0, v2.13.1, v2.13.2, v2.14.0, v2.14.1, v2.15.0, v2.15.1, v2.15.2, v2.16.0, v2.17.0, v2.17.1, v2.18.0
-
net-http 0.9.1
-
openssl 4.0.0
-
optparse 0.8.1
-
pp 0.6.3
-
0.6.2 to v0.6.3
-
prism 1.7.0
-
psych 5.3.1
-
resolv 0.7.0
-
stringio 3.2.0
-
3.1.2 to v3.1.3, v3.1.4, v3.1.5, v3.1.6, v3.1.7, v3.1.8, v3.1.9, v3.2.0
-
strscan 3.1.6
-
time 0.4.2
-
0.4.1 to v0.4.2
-
timeout 0.6.0
-
uri 1.1.1
-
weakref 0.1.4
-
0.1.3 to v0.1.4
-
zlib 3.2.2
-
3.2.1 to v3.2.2
The following bundled gems are updated.
-
minitest 6.0.0
-
power_assert 3.0.1
-
rake 13.3.1
-
test-unit 3.7.3
-
3.6.7 to 3.6.8, 3.6.9, 3.7.0, 3.7.1, 3.7.2, 3.7.3, 3.7.4, 3.7.5
-
rexml 3.4.4
-
rss 0.3.2
-
0.3.1 to 0.3.2
-
net-ftp 0.3.9
-
0.3.8 to v0.3.9
-
net-imap 0.6.2
-
0.5.8 to v0.5.9, v0.5.10, v0.5.11, v0.5.12, v0.5.13, v0.6.0, v0.6.1, v0.6.2
-
net-smtp 0.5.1
-
0.5.0 to v0.5.1
-
matrix 0.4.3
-
0.4.2 to v0.4.3
-
prime 0.1.4
-
0.1.3 to v0.1.4
-
rbs 3.10.0
-
3.8.0 to v3.8.1, v3.9.0.dev.1, v3.9.0.pre.1, v3.9.0.pre.2, v3.9.0, v3.9.1, v3.9.2, v3.9.3, v3.9.4, v3.9.5, v3.10.0.pre.1, v3.10.0.pre.2, v3.10.0
-
typeprof 0.31.1
-
debug 1.11.1
-
1.11.0 to v1.11.1
-
base64 0.3.0
-
0.2.0 to v0.3.0
-
bigdecimal 4.0.1
-
3.1.8 to v3.2.0, v3.2.1, v3.2.2, v3.2.3, v3.3.0, v3.3.1, v4.0.0, v4.0.1
-
drb 2.2.3
-
2.2.1 to v2.2.3
-
syslog 0.3.0
-
0.2.0 to v0.3.0
-
csv 3.3.5
-
repl_type_completor 0.1.12
RubyGems and Bundler
Ruby 4.0 bundled RubyGems and Bundler version 4. see the following links for details.
- Upgrading to RubyGems/Bundler 4 - RubyGems Blog
- 4.0.0 Released - RubyGems Blog
- 4.0.1 Released - RubyGems Blog
- 4.0.2 Released - RubyGems Blog
- 4.0.3 Released - RubyGems Blog
Supported platforms
Windows
- Dropped support for MSVC versions older than 14.0 (_MSC_VER 1900). This means Visual Studio 2015 or later is now required.
Compatibility issues
The following methods were removed from Ractor due to the addition of Ractor::Port:
Ractor.yieldRactor#takeRactor#close_incomingRactor#close_outgoing
[Feature #21262]
ObjectSpace._id2ref is deprecated. [Feature #15408]
Process::Status#& and Process::Status#>> have been removed. They were deprecated in Ruby 3.3. [Bug #19868]
rb_path_check has been removed. This function was used for $SAFE path checking which was removed in Ruby 2.7, and was already deprecated. [Feature #20971]
A backtrace for ArgumentError of “wrong number of arguments” now include the receiver’s class or module name (e.g., in Foo#bar instead of in bar). [Bug #21698]
Backtraces no longer display internal frames. These methods now appear as if it is in the Ruby source file, consistent with other C-implemented methods. [Bug #20968]
Before:
ruby -e '[1].fetch_values(42)'
<internal:array>:211:in 'Array#fetch': index 42 outside of array bounds: -1...1 (IndexError)
from <internal:array>:211:in 'block in Array#fetch_values'
from <internal:array>:211:in 'Array#map!'
from <internal:array>:211:in 'Array#fetch_values'
from -e:1:in '<main>'
After:
$ ruby -e '[1].fetch_values(42)'
-e:1:in 'Array#fetch_values': index 42 outside of array bounds: -1...1 (IndexError)
from -e:1:in '<main>'
Stdlib compatibility issues
CGI library is removed from the default gems. Now we only provide cgi/escape for the following methods:
CGI.escapeandCGI.unescapeCGI.escapeHTMLandCGI.unescapeHTMLCGI.escapeURIComponentandCGI.unescapeURIComponentCGI.escapeElementandCGI.unescapeElement
[Feature #21258]
With the move of Set from stdlib to core class, set/sorted_set.rb has been removed, and SortedSet is no longer an autoloaded constant. Please install the sorted_set gem and require 'sorted_set' to use SortedSet. [Feature #21287]
Net::HTTP
- The default behavior of automatically setting the
Content-Typeheader toapplication/x-www-form-urlencodedfor requests with a body (e.g.,POST,PUT) when the header was not explicitly set has been removed. If your application relied on this automatic default, your requests will now be sent without a Content-Type header, potentially breaking compatibility with certain servers. [GH-net-http #205]
C API updates
IO
rb_thread_fd_closeis deprecated and now a no-op. If you need to expose file descriptors from C extensions to Ruby code, create anIOinstance usingRUBY_IO_MODE_EXTERNALand userb_io_close(io)to close it (this also interrupts and waits for all pending operations on theIOinstance). Directly closing file descriptors does not interrupt pending operations, and may lead to undefined behaviour. In other words, if twoIOobjects share the same file descriptor, closing one does not affect the other. [Feature #18455]
GVL
rb_thread_call_with_gvlnow works with or without the GVL. This allows gems to avoid checkingruby_thread_has_gvl_p. Please still be diligent about the GVL. [Feature #20750]
Set
A C API for Set has been added. The following methods are supported: [Feature #21459]
rb_set_foreachrb_set_newrb_set_new_caparb_set_lookuprb_set_addrb_set_clearrb_set_deleterb_set_size
Implementation improvements
Class#new(ex.Object.new) is faster in all cases, but especially when passing keyword arguments. This has also been integrated into YJIT and ZJIT. [Feature #21254]- GC heaps of different size pools now grow independently, reducing memory usage when only some pools contain long-lived objects
- GC sweeping is faster on pages of large objects
- “Generic ivar” objects (String, Array,
TypedData, etc.) now use a new internal “fields” object for faster instance variable access - The GC avoids maintaining an internal
id2reftable until it is first used, makingobject_idallocation and GC sweeping faster object_idandhashare faster on Class and Module objects- Larger bignum Integers can remain embedded using variable width allocation
Random,Enumerator::Product,Enumerator::Chain,Addrinfo,StringScanner, and some internal objects are now write-barrier protected, which reduces GC overhead.
Ractor
A lot of work has gone into making Ractors more stable, performant, and usable. These improvements bring Ractor implementation closer to leaving experimental status.
-
Performance improvements
-
Frozen strings and the symbol table internally use a lock-free hash set [Feature #21268]
-
Method cache lookups avoid locking in most cases
-
Class (and generic ivar) instance variable access is faster and avoids locking
-
CPU cache contention is avoided in object allocation by using a per-ractor counter
-
CPU cache contention is avoided in xmalloc/xfree by using a thread-local counter
-
object_idavoids locking in most cases -
Bug fixes and stability
-
Fixed possible deadlocks when combining Ractors and Threads
-
Fixed issues with require and autoload in a Ractor
-
Fixed encoding/transcoding issues across Ractors
-
Fixed race conditions in GC operations and method invalidation
-
Fixed issues with processes forking after starting a Ractor
-
GC allocation counts are now accurate under Ractors
-
Fixed TracePoints not working after GC [Bug #19112]
JIT
-
ZJIT
-
Introduce an experimental method-based JIT compiler. Where available, ZJIT can be enabled at runtime with the
--zjitoption or by callingRubyVM::ZJIT.enable. When building Ruby, Rust 1.85.0 or later is required to include ZJIT support. -
As of Ruby 4.0.0, ZJIT is faster than the interpreter, but not yet as fast as YJIT. We encourage experimentation with ZJIT, but advise against deploying it in production for now.
-
Our goal is to make ZJIT faster than YJIT and production-ready in Ruby 4.1.
-
YJIT
-
RubyVM::YJIT.runtime_stats -
ratio_in_yjitno longer works in the default build. Use--enable-yjit=statsonconfigureto enable it on--yjit-stats. -
Add
invalidate_everythingto default stats, which is incremented when every code is invalidated by TracePoint. -
Add
mem_size:andcall_threshold:options toRubyVM::YJIT.enable. -
RJIT
-
--rjitis removed. We will move the implementation of the third-party JIT API to the ruby/rjit repository.