diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index 6290912..0a80352 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -58,6 +58,8 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
+        with:
+          lfs: true
 
       - run: rustup toolchain install stable --profile minimal --no-self-update
       - uses: Swatinem/rust-cache@v2
diff --git a/core/src/parsers/pyreport/chunks.rs b/core/src/parsers/pyreport/chunks.rs
index ddbfb2c..ab569b6 100644
--- a/core/src/parsers/pyreport/chunks.rs
+++ b/core/src/parsers/pyreport/chunks.rs
@@ -46,6 +46,8 @@ use serde::{de, Deserialize};
 use super::report_json::ParsedReportJson;
 use super::utils;
 use crate::error::CodecovError;
+#[cfg(doc)]
+use crate::report::models;
 use crate::report::pyreport::types::{
     self, CoverageType, MissingBranch, Partial, PyreportCoverage, ReportLine,
 };
diff --git a/core/src/parsers/pyreport/mod.rs b/core/src/parsers/pyreport/mod.rs
index baee7f0..3b07289 100644
--- a/core/src/parsers/pyreport/mod.rs
+++ b/core/src/parsers/pyreport/mod.rs
@@ -3,6 +3,8 @@ use std::fs::File;
 use memmap2::Mmap;
 
 use crate::error::Result;
+#[cfg(doc)]
+use crate::report::models;
 use crate::report::SqliteReportBuilder;
 
 pub mod chunks;
@@ -18,14 +20,14 @@ mod utils;
 /// - Chunks file, which describes line-by-line coverage data for each file
 ///
 /// The parser for the report JSON inserts a
-/// [`crate::report::models::SourceFile`] for each file
-/// and a [`crate::report::models::Context`] for each session. It returns two
+/// [`SourceFile`](models::SourceFile) for each file
+/// and a [`Context`](models::Context) for each session. It returns two
 /// hashmaps: one which maps each file's "chunk index" to the database PK for
-/// the `SourceFile` that was inserted for it, and one which maps each session's
+/// the [`SourceFile`](models::SourceFile) that was inserted for it, and one which maps each session's
 /// "session_id" to the database PK for the `Context` that was inserted for it.
 ///
 /// The parser for the chunks file inserts a
-/// [`crate::report::models::CoverageSample`] (and possibly other records) for
+/// [`CoverageSample`](models::CoverageSample) (and possibly other records) for
 /// each coverage measurement contained in the chunks file. It uses the
 /// results of the report JSON parser to figure out the appropriate FKs to
 /// associate a measurement with its `SourceFile` and `Context`(s).