diff --git a/studio/backend/core/inference/diffusion_families.py b/studio/backend/core/inference/diffusion_families.py
index 4522215670..006e34bf18 100644
--- a/studio/backend/core/inference/diffusion_families.py
+++ b/studio/backend/core/inference/diffusion_families.py
@@ -319,6 +319,12 @@ _FAMILIES: tuple[DiffusionFamily, ...] = (
         pipeline_class = "Lumina2Pipeline",
         transformer_class = "Lumina2Transformer2DModel",
         base_repo = "Alpha-VLLM/Lumina-Image-2.0",
+        # Gate-validated hosted checkpoints (28/28 pairs each; LPIPS mean 0.146 int8 /
+        # 0.116 fp8 vs same-seed bf16).
+        prequant_repos = (
+            ("int8", "unsloth/Lumina-Image-2.0-FP8"),
+            ("fp8", "unsloth/Lumina-Image-2.0-FP8"),
+        ),
         aliases = ("lumina-image-2.0", "lumina-image-2", "lumina2"),
         # Published and validated bf16-only upstream; keep the fp16 fallback off like z-image.
         fp16_incompatible = True,
diff --git a/studio/backend/tests/test_diffusion_more_families.py b/studio/backend/tests/test_diffusion_more_families.py
index f3f2e9c291..a65b06c1c5 100644
--- a/studio/backend/tests/test_diffusion_more_families.py
+++ b/studio/backend/tests/test_diffusion_more_families.py
@@ -124,6 +124,15 @@ def test_lumina2_generation_defaults():
     assert default_generation_params("Alpha-VLLM/Lumina-Image-2.0") == (50, 4.0)
 
 
+def test_lumina2_prequant_wiring():
+    # Hosted int8/fp8 checkpoints (gate-validated) serve the family default base.
+    from core.inference.diffusion_families import family_prequant_repo
+
+    fam = detect_family("Alpha-VLLM/Lumina-Image-2.0")
+    for scheme in ("int8", "fp8"):
+        assert family_prequant_repo(fam, scheme) == "unsloth/Lumina-Image-2.0-FP8"
+
+
 def test_lumina2_bf16_component_table_present():
     fam = detect_family("Alpha-VLLM/Lumina-Image-2.0")
     sizes = family_bf16_components_gb(fam)