diff --git a/video/decode/hw_videotoolbox.c b/video/decode/hw_videotoolbox.c
index 8edffb5453..8d8c165a84 100644
--- a/video/decode/hw_videotoolbox.c
+++ b/video/decode/hw_videotoolbox.c
@@ -22,6 +22,7 @@
 
 #include "common/av_common.h"
 #include "common/msg.h"
+#include "options/options.h"
 #include "video/mp_image.h"
 #include "video/decode/lavc.h"
 #include "video/mp_image_pool.h"
@@ -96,10 +97,17 @@ static void print_videotoolbox_error(struct mp_log *log, int lev, char *message,
     mp_msg(log, lev, "%s: %d\n", message, error_code);
 }
 
-static int init_decoder_common(struct lavc_ctx *ctx, int w, int h, AVVideotoolboxContext *vtctx)
+static int init_decoder(struct lavc_ctx *ctx, int w, int h)
 {
     av_videotoolbox_default_free(ctx->avctx);
 
+    AVVideotoolboxContext *vtctx = av_videotoolbox_alloc_context();
+    if (!vtctx)
+        return -1;
+
+    vtctx->cv_pix_fmt_type =
+        mp_imgfmt_to_cvpixelformat(ctx->opts->videotoolbox_format);
+
     int err = av_videotoolbox_default_init2(ctx->avctx, vtctx);
     if (err < 0) {
         print_videotoolbox_error(ctx->log, MSGL_ERR, "failed to init videotoolbox decoder", err);
@@ -109,20 +117,6 @@ static int init_decoder_common(struct lavc_ctx *ctx, int w, int h, AVVideotoolbo
     return 0;
 }
 
-static int init_decoder(struct lavc_ctx *ctx, int w, int h)
-{
-    AVVideotoolboxContext *vtctx = av_videotoolbox_alloc_context();
-    struct mp_vt_ctx *vt = hwdec_devices_load(ctx->hwdec_devs, HWDEC_VIDEOTOOLBOX);
-    vtctx->cv_pix_fmt_type = vt->get_vt_fmt(vt);
-
-    return init_decoder_common(ctx, w, h, vtctx);
-}
-
-static int init_decoder_copy(struct lavc_ctx *ctx, int w, int h)
-{
-    return init_decoder_common(ctx, w, h, NULL);
-}
-
 static void uninit(struct lavc_ctx *ctx)
 {
     if (ctx->avctx)
@@ -179,7 +173,7 @@ const struct vd_lavc_hwdec mp_vd_lavc_videotoolbox_copy = {
     .probe = probe_copy,
     .init = init,
     .uninit = uninit,
-    .init_decoder = init_decoder_copy,
+    .init_decoder = init_decoder,
     .process_image = copy_image,
     .delay_queue = HWDEC_DELAY_QUEUE_COUNT,
 };
diff --git a/video/hwdec.h b/video/hwdec.h
index 9d1035cd6e..f8a089e8c6 100644
--- a/video/hwdec.h
+++ b/video/hwdec.h
@@ -39,7 +39,7 @@ struct mp_hwdec_ctx {
 
     // This is never NULL. Its meaning depends on the .type field:
     //  HWDEC_VDPAU:            struct mp_vaapi_ctx*
-    //  HWDEC_VIDEOTOOLBOX:     struct mp_vt_ctx*
+    //  HWDEC_VIDEOTOOLBOX:     non-NULL dummy pointer
     //  HWDEC_VAAPI:            struct mp_vaapi_ctx*
     //  HWDEC_D3D11VA:          ID3D11Device*
     //  HWDEC_DXVA2:            IDirect3DDevice9*
@@ -64,11 +64,6 @@ struct mp_hwdec_ctx {
                                        struct mp_image_pool *swpool);
 };
 
-struct mp_vt_ctx {
-    void *priv;
-    uint32_t (*get_vt_fmt)(struct mp_vt_ctx *ctx);
-};
-
 // Used to communicate hardware decoder device handles from VO to video decoder.
 struct mp_hwdec_devices;
 
diff --git a/video/out/opengl/hwdec_osx.c b/video/out/opengl/hwdec_osx.c
index ecceab658a..aa350209ec 100644
--- a/video/out/opengl/hwdec_osx.c
+++ b/video/out/opengl/hwdec_osx.c
@@ -27,8 +27,6 @@
 #include "video/mp_image_pool.h"
 #include "video/vt.h"
 #include "hwdec.h"
-#include "common/global.h"
-#include "options/options.h"
 
 struct vt_gl_plane_format {
     GLenum gl_format;
@@ -46,7 +44,6 @@ struct vt_format {
 
 struct priv {
     struct mp_hwdec_ctx hwctx;
-    struct mp_vt_ctx vtctx;
 
     CVPixelBufferRef pbuf;
     GLuint gl_planes[MP_MAX_PLANES];
@@ -123,14 +120,6 @@ static bool check_hwdec(struct gl_hwdec *hw)
     return true;
 }
 
-static uint32_t get_vt_fmt(struct mp_vt_ctx *vtctx)
-{
-    struct gl_hwdec *hw = vtctx->priv;
-    struct vt_format *f =
-        vt_get_gl_format_from_imgfmt(hw->global->opts->videotoolbox_format);
-    return f ? f->cvpixfmt : (uint32_t)-1;
-}
-
 static int create(struct gl_hwdec *hw)
 {
     if (!check_hwdec(hw))
@@ -141,14 +130,10 @@ static int create(struct gl_hwdec *hw)
 
     hw->gl->GenTextures(MP_MAX_PLANES, p->gl_planes);
 
-    p->vtctx = (struct mp_vt_ctx){
-        .priv = hw,
-        .get_vt_fmt = get_vt_fmt,
-    };
     p->hwctx = (struct mp_hwdec_ctx){
         .type = HWDEC_VIDEOTOOLBOX,
         .download_image = mp_vt_download_image,
-        .ctx = &p->vtctx,
+        .ctx = &p->hwctx,
     };
     hwdec_devices_add(hw->devs, &p->hwctx);