diff --git a/src/compressed_tensors/quantization/lifecycle/forward.py b/src/compressed_tensors/quantization/lifecycle/forward.py
index 5b614f35..d685c0c0 100644
--- a/src/compressed_tensors/quantization/lifecycle/forward.py
+++ b/src/compressed_tensors/quantization/lifecycle/forward.py
@@ -20,7 +20,7 @@
 from compressed_tensors.quantization.cache import QuantizedKVParameterCache
 from compressed_tensors.quantization.observers.helpers import (
     calculate_range,
-    compute_memoryless_zp_and_scales,
+    compute_dynamic_scales_and_zp,
 )
 from compressed_tensors.quantization.quant_args import (
     QuantizationArgs,
@@ -380,7 +380,7 @@ def maybe_calibrate_or_quantize(
 
     if args.dynamic:
         # dynamic quantization - no need to invoke observer
-        scale, zero_point = compute_memoryless_zp_and_scales(value=value, args=args)
+        scale, zero_point = compute_dynamic_scales_and_zp(value=value, args=args)
     else:
         # static quantization - get previous scale and zero point from layer
         scale = getattr(module, f"{base_name}_scale")
diff --git a/src/compressed_tensors/quantization/observers/helpers.py b/src/compressed_tensors/quantization/observers/helpers.py
index b7302510..875a05b3 100644
--- a/src/compressed_tensors/quantization/observers/helpers.py
+++ b/src/compressed_tensors/quantization/observers/helpers.py
@@ -29,13 +29,14 @@
     "calculate_qparams",
     "get_observer_token_count",
     "calculate_range",
-    "compute_memoryless_zp_and_scales",
+    "compute_dynamic_scales_and_zp",
 ]
 
 
-def compute_memoryless_zp_and_scales(value: Tensor, args: QuantizationArgs):
+def compute_dynamic_scales_and_zp(value: Tensor, args: QuantizationArgs):
     """
-    Returns the min and max values of observed tensor
+    Returns the computed scales and zero points for dynamic activation
+    qunatization.
 
     :param value: tensor to calculate quantization parameters for
     :param args: quantization args
diff --git a/src/compressed_tensors/quantization/quant_args.py b/src/compressed_tensors/quantization/quant_args.py
index 920cd99b..339a2a66 100644
--- a/src/compressed_tensors/quantization/quant_args.py
+++ b/src/compressed_tensors/quantization/quant_args.py
@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import warnings
 from enum import Enum
 from typing import Any, Dict, Optional, Union
 
@@ -171,6 +172,8 @@ def validate_model_after(model: "QuantizationArgs") -> Dict[str, Any]:
         strategy = model.strategy
         group_size = model.group_size
         actorder = model.actorder
+        dynamic = model.dynamic
+        observer = model.observer
 
         # infer strategy
         if strategy is None:
@@ -207,6 +210,27 @@ def validate_model_after(model: "QuantizationArgs") -> Dict[str, Any]:
                 "activation ordering"
             )
 
+        # if we have not set an observer and we
+        # are running static quantization, use minmax
+        if not observer and not dynamic:
+            model.observer = "minmax"
+
+        if dynamic:
+            if strategy not in (
+                QuantizationStrategy.TOKEN,
+                QuantizationStrategy.TENSOR,
+            ):
+                raise ValueError(
+                    f"One of {QuantizationStrategy.TOKEN} or "
+                    f"{QuantizationStrategy.TENSOR} must be used for dynamic ",
+                    "quantization",
+                )
+            if observer is not None:
+                warnings.warn(
+                    "No observer is used for dynamic quantization, setting to None"
+                )
+                model.observer = None
+
         # write back modified values
         model.strategy = strategy
         return model