diff --git a/PhysicsTools/ONNXRuntime/BuildFile.xml b/PhysicsTools/ONNXRuntime/BuildFile.xml
index 74335c38157af..28c17eb4f0273 100644
--- a/PhysicsTools/ONNXRuntime/BuildFile.xml
+++ b/PhysicsTools/ONNXRuntime/BuildFile.xml
@@ -1,5 +1,7 @@
 <use name="onnxruntime"/>
 <use name="FWCore/Utilities"/>
+<use name="HeterogeneousCore/CUDAServices"/>
+<use name="FWCore/ServiceRegistry"/>
 <export>
   <lib name="1"/>
 </export>
diff --git a/PhysicsTools/ONNXRuntime/interface/ONNXSessionOptions.h b/PhysicsTools/ONNXRuntime/interface/ONNXSessionOptions.h
new file mode 100644
index 0000000000000..f43bc739d11dc
--- /dev/null
+++ b/PhysicsTools/ONNXRuntime/interface/ONNXSessionOptions.h
@@ -0,0 +1,32 @@
+#ifndef PHYSICSTOOLS_ONNXRUNTIME_ONNXSESSIONOPTIONS_H
+#define PHYSICSTOOLS_ONNXRUNTIME_ONNXSESSIONOPTIONS_H
+
+#include "HeterogeneousCore/CUDAServices/interface/CUDAService.h"
+#include "FWCore/ServiceRegistry/interface/Service.h"
+#include "PhysicsTools/ONNXRuntime/interface/ONNXRuntime.h"
+#include "onnxruntime/core/session/onnxruntime_cxx_api.h"
+#include <string>
+
+namespace cms::Ort {
+
+  // param_backend
+  //    cpu -> Use CPU backend
+  //    cuda -> Use cuda backend
+  //    default -> Use best available
+  inline ::Ort::SessionOptions getSessionOptions(const std::string &param_backend) {
+    auto backend = cms::Ort::Backend::cpu;
+    if (param_backend == "cuda")
+      backend = cms::Ort::Backend::cuda;
+
+    if (param_backend == "default") {
+      edm::Service<CUDAService> cs;
+      if (cs.isAvailable() and cs->enabled()) {
+        backend = cms::Ort::Backend::cuda;
+      }
+    }
+
+    return ONNXRuntime::defaultSessionOptions(backend);
+  }
+}  // namespace cms::Ort
+
+#endif
diff --git a/PhysicsTools/ONNXRuntime/src/ONNXRuntime.cc b/PhysicsTools/ONNXRuntime/src/ONNXRuntime.cc
index 130e4544585b9..60ddb44b6e436 100644
--- a/PhysicsTools/ONNXRuntime/src/ONNXRuntime.cc
+++ b/PhysicsTools/ONNXRuntime/src/ONNXRuntime.cc
@@ -20,7 +20,11 @@ namespace cms::Ort {
 
   using namespace ::Ort;
 
+#ifdef ONNXDebug
+  const Env ONNXRuntime::env_(ORT_LOGGING_LEVEL_INFO, "");
+#else
   const Env ONNXRuntime::env_(ORT_LOGGING_LEVEL_ERROR, "");
+#endif
 
   ONNXRuntime::ONNXRuntime(const std::string& model_path, const SessionOptions* session_options) {
     // create session
@@ -80,10 +84,12 @@ namespace cms::Ort {
     SessionOptions sess_opts;
     sess_opts.SetIntraOpNumThreads(1);
     if (backend == Backend::cuda) {
-      // https://www.onnxruntime.ai/docs/reference/execution-providers/CUDA-ExecutionProvider.html
       OrtCUDAProviderOptions options;
       sess_opts.AppendExecutionProvider_CUDA(options);
     }
+#ifdef ONNX_PROFILE
+    sess_opts.EnableProfiling("ONNXProf");
+#endif
     return sess_opts;
   }
 
@@ -140,6 +146,7 @@ namespace cms::Ort {
     }
 
     // run
+
     auto output_tensors = session_->Run(RunOptions{nullptr},
                                         input_node_names_.data(),
                                         input_tensors.data(),
diff --git a/RecoBTag/ONNXRuntime/plugins/BoostedJetONNXJetTagsProducer.cc b/RecoBTag/ONNXRuntime/plugins/BoostedJetONNXJetTagsProducer.cc
index 5cdf4ed1ee962..d51d2811be1a0 100644
--- a/RecoBTag/ONNXRuntime/plugins/BoostedJetONNXJetTagsProducer.cc
+++ b/RecoBTag/ONNXRuntime/plugins/BoostedJetONNXJetTagsProducer.cc
@@ -14,7 +14,7 @@
 #include "DataFormats/BTauReco/interface/DeepBoostedJetTagInfo.h"
 
 #include "PhysicsTools/ONNXRuntime/interface/ONNXRuntime.h"
-
+#include "PhysicsTools/ONNXRuntime/interface/ONNXSessionOptions.h"
 #include "RecoBTag/FeatureTools/interface/deep_helpers.h"
 
 #include <iostream>
@@ -126,12 +126,20 @@ void BoostedJetONNXJetTagsProducer::fillDescriptions(edm::ConfigurationDescripti
                                          "probQCDothers",
                                      });
   desc.addOptionalUntracked<bool>("debugMode", false);
+  desc.add<std::string>("onnx_backend", "default");
 
   descriptions.addWithDefaultLabel(desc);
 }
 
 std::unique_ptr<ONNXRuntime> BoostedJetONNXJetTagsProducer::initializeGlobalCache(const edm::ParameterSet &iConfig) {
-  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath());
+  std::string backend = iConfig.getParameter<std::string>("onnx_backend");
+
+  auto session_options = cms::Ort::getSessionOptions(backend);
+  // Sept 8, 2022 - on gpu, this model crashes with all optimizations on
+  if (backend != "cpu")
+    session_options.SetGraphOptimizationLevel(GraphOptimizationLevel::ORT_ENABLE_BASIC);
+  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath(),
+                                       &session_options);
 }
 
 void BoostedJetONNXJetTagsProducer::globalEndJob(const ONNXRuntime *cache) {}
diff --git a/RecoBTag/ONNXRuntime/plugins/DeepCombinedONNXJetTagsProducer.cc b/RecoBTag/ONNXRuntime/plugins/DeepCombinedONNXJetTagsProducer.cc
index 99193460081a7..d1a43103702d1 100644
--- a/RecoBTag/ONNXRuntime/plugins/DeepCombinedONNXJetTagsProducer.cc
+++ b/RecoBTag/ONNXRuntime/plugins/DeepCombinedONNXJetTagsProducer.cc
@@ -14,7 +14,7 @@
 #include "DataFormats/BTauReco/interface/DeepFlavourTagInfo.h"
 
 #include "PhysicsTools/ONNXRuntime/interface/ONNXRuntime.h"
-
+#include "PhysicsTools/ONNXRuntime/interface/ONNXSessionOptions.h"
 #include "RecoBTag/ONNXRuntime/interface/tensor_fillers.h"
 #include "RecoBTag/ONNXRuntime/interface/tensor_configs.h"
 
@@ -136,12 +136,15 @@ void DeepCombinedONNXJetTagsProducer::fillDescriptions(edm::ConfigurationDescrip
   desc.add<std::vector<std::string>>("flav_names", std::vector<std::string>{"probb", "probc", "probuds", "probg"});
   desc.add<double>("min_jet_pt", 15.0);
   desc.add<double>("max_jet_eta", 2.5);
+  desc.add<std::string>("onnx_backend", "default");
 
   descriptions.add("pfDeepCombinedJetTags", desc);
 }
 
 std::unique_ptr<ONNXRuntime> DeepCombinedONNXJetTagsProducer::initializeGlobalCache(const edm::ParameterSet& iConfig) {
-  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath());
+  auto session_options = cms::Ort::getSessionOptions(iConfig.getParameter<std::string>("onnx_backend"));
+  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath(),
+                                       &session_options);
 }
 
 void DeepCombinedONNXJetTagsProducer::globalEndJob(const ONNXRuntime* cache) {}
diff --git a/RecoBTag/ONNXRuntime/plugins/DeepDoubleXONNXJetTagsProducer.cc b/RecoBTag/ONNXRuntime/plugins/DeepDoubleXONNXJetTagsProducer.cc
index 2c73eaac090fc..c6360d4f59cab 100644
--- a/RecoBTag/ONNXRuntime/plugins/DeepDoubleXONNXJetTagsProducer.cc
+++ b/RecoBTag/ONNXRuntime/plugins/DeepDoubleXONNXJetTagsProducer.cc
@@ -14,7 +14,7 @@
 #include "DataFormats/BTauReco/interface/DeepDoubleXTagInfo.h"
 
 #include "PhysicsTools/ONNXRuntime/interface/ONNXRuntime.h"
-
+#include "PhysicsTools/ONNXRuntime/interface/ONNXSessionOptions.h"
 #include <algorithm>
 #include <iostream>
 #include <fstream>
@@ -121,6 +121,8 @@ void DeepDoubleXONNXJetTagsProducer::fillDescriptions(edm::ConfigurationDescript
            "CvB" >> (PDPSD("flav_names", std::vector<std::string>{"probHbb", "probHcc"}, true) and
                      PDFIP("model_path", FIP("RecoBTag/Combined/data/DeepDoubleX/94X/V01/DDCvB.onnx"), true));
   };
+  desc.add<std::string>("onnx_backend", "default");
+
   auto descBvL(desc);
   descBvL.ifValue(edm::ParameterDescription<std::string>("flavor", "BvL", true), flavorCases());
   descriptions.add("pfDeepDoubleBvLJetTags", descBvL);
@@ -135,7 +137,9 @@ void DeepDoubleXONNXJetTagsProducer::fillDescriptions(edm::ConfigurationDescript
 }
 
 std::unique_ptr<ONNXRuntime> DeepDoubleXONNXJetTagsProducer::initializeGlobalCache(const edm::ParameterSet& iConfig) {
-  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath());
+  auto session_options = cms::Ort::getSessionOptions(iConfig.getParameter<std::string>("onnx_backend"));
+  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath(),
+                                       &session_options);
 }
 
 void DeepDoubleXONNXJetTagsProducer::globalEndJob(const ONNXRuntime* cache) {}
diff --git a/RecoBTag/ONNXRuntime/plugins/DeepFlavourONNXJetTagsProducer.cc b/RecoBTag/ONNXRuntime/plugins/DeepFlavourONNXJetTagsProducer.cc
index 10dacfcd5e819..7fe7b96cb3007 100644
--- a/RecoBTag/ONNXRuntime/plugins/DeepFlavourONNXJetTagsProducer.cc
+++ b/RecoBTag/ONNXRuntime/plugins/DeepFlavourONNXJetTagsProducer.cc
@@ -14,6 +14,7 @@
 #include "DataFormats/BTauReco/interface/DeepFlavourTagInfo.h"
 
 #include "PhysicsTools/ONNXRuntime/interface/ONNXRuntime.h"
+#include "PhysicsTools/ONNXRuntime/interface/ONNXSessionOptions.h"
 
 using namespace cms::Ort;
 
@@ -84,12 +85,15 @@ void DeepFlavourONNXJetTagsProducer::fillDescriptions(edm::ConfigurationDescript
   desc.add<std::vector<std::string>>("output_names", {"ID_pred/Softmax:0"});
   desc.add<std::vector<std::string>>(
       "flav_names", std::vector<std::string>{"probb", "probbb", "problepb", "probc", "probuds", "probg"});
+  desc.add<std::string>("onnx_backend", "default");
 
   descriptions.add("pfDeepFlavourJetTags", desc);
 }
 
 std::unique_ptr<ONNXRuntime> DeepFlavourONNXJetTagsProducer::initializeGlobalCache(const edm::ParameterSet& iConfig) {
-  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath());
+  auto session_options = cms::Ort::getSessionOptions(iConfig.getParameter<std::string>("onnx_backend"));
+  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath(),
+                                       &session_options);
 }
 
 void DeepFlavourONNXJetTagsProducer::globalEndJob(const ONNXRuntime* cache) {}
diff --git a/RecoBTag/ONNXRuntime/plugins/DeepVertexONNXJetTagsProducer.cc b/RecoBTag/ONNXRuntime/plugins/DeepVertexONNXJetTagsProducer.cc
index c9805e66f04f7..60b975d33fbc9 100644
--- a/RecoBTag/ONNXRuntime/plugins/DeepVertexONNXJetTagsProducer.cc
+++ b/RecoBTag/ONNXRuntime/plugins/DeepVertexONNXJetTagsProducer.cc
@@ -14,6 +14,7 @@
 #include "DataFormats/BTauReco/interface/DeepFlavourTagInfo.h"
 
 #include "PhysicsTools/ONNXRuntime/interface/ONNXRuntime.h"
+#include "PhysicsTools/ONNXRuntime/interface/ONNXSessionOptions.h"
 
 #include "RecoBTag/ONNXRuntime/interface/tensor_fillers.h"
 #include "RecoBTag/ONNXRuntime/interface/tensor_configs.h"
@@ -111,12 +112,15 @@ void DeepVertexONNXJetTagsProducer::fillDescriptions(edm::ConfigurationDescripti
   desc.add<std::vector<std::string>>("flav_names", std::vector<std::string>{"probb", "probc", "probuds", "probg"});
   desc.add<double>("min_jet_pt", 15.0);
   desc.add<double>("max_jet_eta", 2.5);
+  desc.add<std::string>("onnx_backend", "default");
 
   descriptions.add("pfDeepVertexJetTags", desc);
 }
 
 std::unique_ptr<ONNXRuntime> DeepVertexONNXJetTagsProducer::initializeGlobalCache(const edm::ParameterSet& iConfig) {
-  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath());
+  auto session_options = cms::Ort::getSessionOptions(iConfig.getParameter<std::string>("onnx_backend"));
+  return std::make_unique<ONNXRuntime>(iConfig.getParameter<edm::FileInPath>("model_path").fullPath(),
+                                       &session_options);
 }
 
 void DeepVertexONNXJetTagsProducer::globalEndJob(const ONNXRuntime* cache) {}
diff --git a/RecoParticleFlow/PFProducer/plugins/MLPFProducer.cc b/RecoParticleFlow/PFProducer/plugins/MLPFProducer.cc
index afd449922898a..cc791e464d251 100644
--- a/RecoParticleFlow/PFProducer/plugins/MLPFProducer.cc
+++ b/RecoParticleFlow/PFProducer/plugins/MLPFProducer.cc
@@ -5,6 +5,7 @@
 
 #include "DataFormats/ParticleFlowCandidate/interface/PFCandidate.h"
 #include "PhysicsTools/ONNXRuntime/interface/ONNXRuntime.h"
+#include "PhysicsTools/ONNXRuntime/interface/ONNXSessionOptions.h"
 #include "RecoParticleFlow/PFProducer/interface/MLPFModel.h"
 
 #include "DataFormats/ParticleFlowReco/interface/PFBlockElementTrack.h"
@@ -160,7 +161,8 @@ void MLPFProducer::produce(edm::Event& event, const edm::EventSetup& setup) {
 }
 
 std::unique_ptr<ONNXRuntime> MLPFProducer::initializeGlobalCache(const edm::ParameterSet& params) {
-  return std::make_unique<ONNXRuntime>(params.getParameter<edm::FileInPath>("model_path").fullPath());
+  auto session_options = cms::Ort::getSessionOptions(params.getParameter<std::string>("onnx_backend"));
+  return std::make_unique<ONNXRuntime>(params.getParameter<edm::FileInPath>("model_path").fullPath(), &session_options);
 }
 
 void MLPFProducer::globalEndJob(const ONNXRuntime* cache) {}
@@ -173,6 +175,7 @@ void MLPFProducer::fillDescriptions(edm::ConfigurationDescriptions& descriptions
       edm::FileInPath(
           "RecoParticleFlow/PFProducer/data/mlpf/"
           "mlpf_2021_11_16__no_einsum__all_data_cms-best-of-asha-scikit_20211026_042043_178263.workergpu010.onnx"));
+  desc.add<std::string>("onnx_backend", "default");
   descriptions.addWithDefaultLabel(desc);
 }