diff --git a/core/dr-inference-engine/src/session.rs b/core/dr-inference-engine/src/session.rs
index 6b70373..ff2cc54 100644
--- a/core/dr-inference-engine/src/session.rs
+++ b/core/dr-inference-engine/src/session.rs
@@ -1,6 +1,5 @@
 //! One session builder per rung (docs/inference.md §2, §7, §9).
 
-use ort::session::builder::GraphOptimizationLevel;
 use ort::session::Session;
 
 use crate::{Config, Role, Rung};
@@ -14,9 +13,12 @@ use crate::{Config, Role, Rung};
 /// its clock instead (§4): a provider that hands real work to the CPU is
 /// slower than the CPU floor and rejected by the same measurement.
 pub fn build(rung: Rung, role: Role, bytes: &[u8], cfg: &Config) -> ort::Result<Session> {
-    let mut b = Session::builder()?
-        .with_optimization_level(GraphOptimizationLevel::Level3)?
-        .with_intra_threads(threads(cfg))?;
+    // No optimisation level named. ONNX Runtime's default is already its
+    // fullest, and on tract any level but "disabled" means `into_optimized`,
+    // whose optimiser divides by zero inside yolo26n-seg (tract-data
+    // `stack_tensors`) — a panic across the C API, which is an abort. The
+    // app never asked tract for that and does not start now.
+    let mut b = Session::builder()?.with_intra_threads(threads(cfg))?;
     // A Hexagon session loads the compiled context when there is one and
     // compiles it from the model when there is not; the engine thread is
     // what makes the second case rare (§6).