diff --git a/runs/frontier-multicase-sufficiency-v0/best_effort/fixed_cohort_rank.py b/runs/frontier-multicase-sufficiency-v0/best_effort/fixed_cohort_rank.py
index 95ab0e2..cc1cf73 100644
--- a/runs/frontier-multicase-sufficiency-v0/best_effort/fixed_cohort_rank.py
+++ b/runs/frontier-multicase-sufficiency-v0/best_effort/fixed_cohort_rank.py
@@ -593,15 +593,14 @@ def run_frontier(args: argparse.Namespace) -> Path:
     audit_protocol(args.protocol_manifest)
     profile_paths = frontier_grid.resolve_profile_paths(args.profile_root)
     profile_contract = _validate_profile_contract(profile_paths)
-    cache_equivalence = json.loads(args.cache_equivalence_artifact.read_text())
+    ep_equivalence = json.loads(args.ep_equivalence_artifact.read_text())
     if (
-        cache_equivalence.get("schema")
-        != "frontier-ep-critical-lane-cache-equivalence-v1"
-        or not cache_equivalence.get("byte_identical")
+        ep_equivalence.get("schema") != "frontier-ep-batched-lane-equivalence-v1"
+        or not ep_equivalence.get("byte_identical")
     ):
-        raise ValueError("EP critical-lane cache lacks byte-identical evidence")
-    if Path(cache_equivalence["new_source"]).resolve() != args.frontier_source.resolve():
-        raise ValueError("cache equivalence was measured on a different Frontier source")
+        raise ValueError("batched EP lane prediction lacks byte-identical evidence")
+    if Path(ep_equivalence["new_source"]).resolve() != args.frontier_source.resolve():
+        raise ValueError("EP equivalence was measured on a different Frontier source")
     configs = frontier_grid.selected_configs(args.config)
     args.output_root.mkdir(parents=True, exist_ok=True)
     run_manifest = {
@@ -625,11 +624,11 @@ def run_frontier(args: argparse.Namespace) -> Path:
                 for name, path in profile_paths.items()
             },
         },
-        "ep_critical_lane_cache_equivalence": {
-            "path": str(args.cache_equivalence_artifact.resolve()),
-            "sha256": sha256(args.cache_equivalence_artifact),
+        "ep_prediction_optimization_equivalence": {
+            "path": str(args.ep_equivalence_artifact.resolve()),
+            "sha256": sha256(args.ep_equivalence_artifact),
             "request_metrics_byte_identical": True,
-            "request_metrics_sha256": cache_equivalence["new_sha256"],
+            "request_metrics_sha256": ep_equivalence["new_sha256"],
         },
         "kv_capacity_evidence": {
             "tp4": {
@@ -649,6 +648,7 @@ def run_frontier(args: argparse.Namespace) -> Path:
             "execution_profiles": "community_vllm_0.10.2_serving_entrypoints",
             "moe_ep_prefill": "global_routing_then_slowest_local_ep_lane",
             "moe_ep_critical_lane_cache": "exact_(total_routed_tokens,layer_id)_memoization",
+            "moe_ep_lane_inference": "eight_independent_rows_in_one_exact_forest_predict_call",
             "kv_blocks": "measured_community_vllm_and_fixed_per_topology",
             "cpu_overhead_modeling": "disabled_no_native_community_vllm_records",
             "end_to_end_action_specific_calibration": False,
@@ -1458,7 +1458,7 @@ def parse_args() -> argparse.Namespace:
     frontier.add_argument("--output-root", type=Path, required=True)
     frontier.add_argument("--tp4-capacity-artifact", type=Path, required=True)
     frontier.add_argument("--tp8-capacity-artifact", type=Path, required=True)
-    frontier.add_argument("--cache-equivalence-artifact", type=Path, required=True)
+    frontier.add_argument("--ep-equivalence-artifact", type=Path, required=True)
     frontier.add_argument("--max-parallel-configs", type=int, default=1)
     frontier.add_argument("--config", action="append")