diff --git a/vllm_spyre/v1/worker/spyre_model_runner.py b/vllm_spyre/v1/worker/spyre_model_runner.py
index be57fee6c..8e051c808 100644
--- a/vllm_spyre/v1/worker/spyre_model_runner.py
+++ b/vllm_spyre/v1/worker/spyre_model_runner.py
@@ -308,8 +308,17 @@ def build_input_batch(self) -> SamplingInputBatch:
         # TODO(Max): logits processor list should be extensible via engine
         # constructor argument; for now the list is fixed to builtin processors
         logits_processors = get_builtin_logits_processors(self.vllm_config)
+
+        # fix for fp8 batch size 1 (continuous batching): set input batch to
+        # fit at minimum 2 requests for the dynamic warmup and then reset input
+        # batch to fit max_num_seqs requests after warmup has completed
+        min_seqs_required = 2 if (envs_spyre.VLLM_SPYRE_USE_CB
+                                  and self.model_config.quantization
+                                  and self.warmup_mode) else 1
+
         return SamplingInputBatch(
-            max_num_reqs=self.scheduler_config.max_num_seqs,
+            max_num_reqs=max(min_seqs_required,
+                             self.scheduler_config.max_num_seqs),
             max_model_len=self.model_config.max_model_len,
             device=self.device,
             pin_memory=self.pin_memory,
@@ -841,6 +850,9 @@ def pre_warmup(self) -> None:
 
     def complete_warmup(self) -> None:
         super().complete_warmup()
+        # Fix for fp8 batch size 1: update the input_batch after warmup
+        if self.model_config.quantization:
+            self.input_batch = self.build_input_batch()
         # get the number or pages from the actual Spyre card after the warmup
         # and set it accordingly in the model runner and for the kv cache size
         n_blocks_avail = self.model.model.get_num_blocks_available()