v5.1.0 (2 September 2025)

• Added functions for the embedded Laplace approximation. (#3097, #3209, #3210)
• Added new overloads for sum_to_zero_constrain and sum_to_zero_free that accept Eigen Matrix types. (#3169)
• The simplex transform is now defined in terms of the isometric log ratio transform instead of stick breaking. (#3171)
• Renamed hypergeometric_1f0 to hypergeometric_1F0 for consistency, fixed compilation for hypergeometric_2f1 with autodiff second argument. (#3165)
• Simplified implementation of chol2inv using matrix functions. (#3136)
• Improved the numerical stability of the gradient calculations for inv_Phi and std_normal_log_qf. (#3139, #3207)
• Simplified vectorisation of std_normal_lpdf. (#3140)
• Removed redundant computation in grad_reg_inc_beta. (#3159)
• Updated signature for apply_scalar_binary to take the functor as its first argument, instead of the last. (#3156)
• Added perfect forwarding to all functions that use the apply_ family of functors. (#3221, #3228, #3229)
• Fixed expression test errors with poisson_binomial_rng, added support for row-vector and std::vector<double> theta arguments for poisson_binomial. (#3178)
• Updated inv_logit to use the logistic function from Eigen. (#3155, #3160)
• Update the constraint functions templating to be more flexible. (#3151)
• Added ARM64 testing to the Github CI for Windows and Ubuntu (#3177, #3173, #3180, #3174, #3158, #3164)
• Disabled Make's implicit rules in our Makefiles. (#3183)
• Updated Boost to version 1.87.0 (#3185)
• Patched an issue preventing TBB from building with clang 21.1 (#3223)