Guile implements Scheme as described in the Revised^5 Report on the Algorithmic Language Scheme (usually known as R5RS), providing clean and general data and control structures. Guile goes beyond the rather austere language presented in R5RS, extending it with a module system, full access to POSIX system calls, networking support, multiple threads, dynamic linking, a foreign function call interface, powerful string processing, and many other features needed for programming in the real world.
In 2007, the Scheme community agreed upon and published R6RS, a significant installment in the RnRS series. R6RS expands the core Scheme language, and standardises many non-core functions that implementations—including Guile—have previously done in different ways. Over time, Guile has been updated to incorporate almost all of the features of R6RS, and to adjust some existing features to conform to the R6RS specification. See R6RS Support, for full details.
In parallel to official standardization efforts, the SRFI process (http://srfi.schemers.org/) standardises interfaces for many practical needs, such as multithreaded programming and multidimensional arrays. Guile supports many SRFIs, as documented in detail in SRFI Support Modules.
The process that led to the R6RS standard brought a split in the Scheme community to the surface. The implementors that wrote R6RS considered that it was impossible to write useful, portable programs in R5RS, and that only an ambitious standard could solve this problem. However, part of the Scheme world saw the R6RS effort as too broad, and as having included some components that would never be adopted by more minimalistic Scheme implementations. This second group succeeded in taking control of the official Scheme standardization track and in 2013 released a more limited R7RS, essentially consisting of R5RS, plus a module system. Guile supports R7RS also. See R7RS Support.
With R6RS and R7RS, the unified Scheme standardization process appears to have more or less run its course. There will continue to be more code written in terms of both systems, and modules defined using the SRFI process, and Guile will support both. However for future directions, Guile takes inspiration from other related language communities: Racket, Clojure, Concurrent ML, and so on.
In summary, Guile supports writing and running code written to the R5RS, R6RS, and R7RS Scheme standards, and also supports a number of SRFI modules. However for most users, until a need for cross-implementation portability has been identified, we recommend using the parts of Guile that are useful in solving the problem at hand, regardless of whether they proceed from a standard or whether they are Guile-specific.