mmc(...)

## Not run: 
# simulate some data
dat <- data.frame(
  y = rnorm(100), x1 = rnorm(100), x2 = rnorm(100),
  g1 = sample(1:10, 100, TRUE), g2 = sample(1:10, 100, TRUE)
)

# multi-membership model with level specific covariate values
dat$xc <- (dat$x1 + dat$x2) / 2 
fit <- brm(y ~ xc + (1 + mmc(x1, x2) | mm(g1, g2)), data = dat)
summary(fit)

## End(Not run)