diff --git a/Project.toml b/Project.toml
index bf474dfe..0c081ab4 100644
--- a/Project.toml
+++ b/Project.toml
@@ -1,7 +1,7 @@
 name = "LuxLib"
 uuid = "82251201-b29d-42c6-8e01-566dec8acb11"
 authors = ["Avik Pal <avikpal@mit.edu> and contributors"]
-version = "0.3.39"
+version = "1.0.0"
 
 [deps]
 ArrayInterface = "4fba245c-0d91-5ea0-9b3e-6abc04ee57a9"
diff --git a/src/LuxLib.jl b/src/LuxLib.jl
index 67796493..766f7642 100644
--- a/src/LuxLib.jl
+++ b/src/LuxLib.jl
@@ -54,8 +54,6 @@ include("impl/forward_diff.jl")
 include("impl/matmul.jl")
 include("impl/normalization.jl")
 
-include("deprecations.jl")
-
 export batchnorm, groupnorm, instancenorm, layernorm, alpha_dropout, dropout
 export fused_dense_bias_activation, fused_conv_bias_activation
 export fast_activation, fast_activation!!
diff --git a/src/deprecations.jl b/src/deprecations.jl
deleted file mode 100644
index 3b002bf4..00000000
--- a/src/deprecations.jl
+++ /dev/null
@@ -1,39 +0,0 @@
-# Deprecations for version 0.4
-## normalization
-@deprecate batchnorm(x, scale, bias, running_mean, running_var, σ::F=identity;
-    momentum::Real, training::Val, epsilon::Real) where {F} batchnorm(
-    x, scale, bias, running_mean, running_var, training, σ, momentum, epsilon)
-
-@deprecate groupnorm(x, scale, bias, σ::F=identity; groups::Int, epsilon::Real) where {F} groupnorm(
-    x, scale, bias, groups, σ, epsilon)
-
-@deprecate instancenorm(x, scale, bias, σ::F=identity; epsilon, training) where {F} instancenorm(
-    x, scale, bias, training, σ, epsilon)
-
-@deprecate layernorm(x, scale, bias, σ::F=identity; dims, epsilon) where {F} layernorm(
-    x, scale, bias, σ, dims, epsilon)
-
-## dropout
-@deprecate dropout(
-    rng::AbstractRNG, x::AbstractArray, p::T, training::Val, invp::T; dims) where {T} dropout(
-    rng, x, p, training, invp, dims)
-
-@deprecate dropout(
-    rng::AbstractRNG, x::AbstractArray, p::T, training::Val; dims, invp::T=inv(p)) where {T} dropout(
-    rng, x, p, training, invp, dims)
-
-@deprecate dropout(rng::AbstractRNG, x::AbstractArray{T1, N}, mask::AbstractArray{T2, N},
-    p::T, training::Val, um::Val, invp::T; dims) where {T, T1, T2, N} dropout(
-    rng, x, mask, p, training, um, invp, dims)
-
-@deprecate dropout(rng::AbstractRNG, x::AbstractArray{T1, N}, mask::AbstractArray{T2, N},
-    p::T, training::Val, um::Val; dims, invp::T=inv(p)) where {T, T1, T2, N} dropout(
-    rng, x, mask, p, training, um, invp, dims)
-
-# bias activation. While this is not public, we used it in Lux
-function __apply_bias_activation(σ::F, x, bias::AbstractArray) where {F}
-    __depwarn("`__apply_bias_activation` is deprecated and will be removed in the next \
-               release. Use `bias_activation` instead.",
-        :__apply_bias_activation)
-    return __bias_activation_impl(σ, x, _vec(bias))
-end