diff --git a/agent/conversation_loop.py b/agent/conversation_loop.py
index 83681e426670..5b4185a24124 100644
--- a/agent/conversation_loop.py
+++ b/agent/conversation_loop.py
@@ -3061,14 +3061,21 @@ def run_conversation(
                 # (``/new``), switch to a larger-context model, or reduce
                 # attachments.  Forced compaction via ``/compress``
                 # (``force=True``) is unaffected — it never reaches this loop.
+                #
+                # Output-cap errors (max_tokens too large) are NOT input
+                # overflow — the recovery is a max_tokens-only retry that
+                # does not require compression.  Exempt them from this guard
+                # so the retry still fires even when compression is disabled.
                 _overflow_reasons = {
                     FailoverReason.long_context_tier,
                     FailoverReason.payload_too_large,
                     FailoverReason.context_overflow,
                 }
+                _is_output_cap_error = is_output_cap_error(error_msg)
                 if (
                     classified.reason in _overflow_reasons
                     and not getattr(agent, "compression_enabled", True)
+                    and not _is_output_cap_error
                 ):
                     agent._flush_status_buffer()
                     agent._vprint(
@@ -3487,8 +3494,10 @@ def run_conversation(
                         if local_available_out > 0:
                             safe_out = max(1, min(available_out, local_available_out) - 64)
                         else:
-                            # Rough local estimate can overshoot; provider truth
-                            # still allows a minimal retry.
+                            # The rough local estimate can overshoot the real
+                            # request size.  Fall back to the provider-reported
+                            # budget, which is authoritative for the failed
+                            # request.
                             safe_out = max(1, available_out - 64)
                         agent._ephemeral_max_output_tokens = safe_out
                         agent._buffer_vprint(