diff --git a/agent/agent_init.py b/agent/agent_init.py
index 495260d2188..f1666a1aafb 100644
--- a/agent/agent_init.py
+++ b/agent/agent_init.py
@@ -77,8 +77,8 @@ def _build_codex_gpt5_autoraise_notice(autoraise: Dict[str, Any]) -> str:
     include the exact opt-back-out command.
     """
     model = str(autoraise.get("model") or "gpt-5.4/5.5").strip().lower().rsplit("/", 1)[-1]
-    # gpt-5.3-codex-spark has a native 128K window; the gpt-5.4/5.5 family is
-    # capped at 272K by the Codex OAuth backend.
+    # gpt-5.3-codex-spark has a native 128K window; the gpt-5.4/5.5/5.6 family
+    # is capped at 272K by the Codex OAuth backend.
     cap = "128K" if model.startswith("gpt-5.3-codex-spark") else "272K"
     from_pct = int(round(autoraise["from"] * 100))
     to_pct = int(round(autoraise["to"] * 100))
diff --git a/agent/auxiliary_client.py b/agent/auxiliary_client.py
index d55f7df4fcd..b1d9a8135e6 100644
--- a/agent/auxiliary_client.py
+++ b/agent/auxiliary_client.py
@@ -314,15 +314,18 @@ def _is_arcee_trinity_thinking(model: Optional[str]) -> bool:
     return bare == "trinity-large-thinking"
 
 
-# Context window enforced by ChatGPT's Codex OAuth backend for gpt-5.4/5.5.
-# The raw OpenAI API and OpenRouter expose 1.05M for the same slug, but the
-# Codex backend hard-caps at 272K (verified live: a ~330K-token request to
+# Context window enforced by ChatGPT's Codex OAuth backend for the
+# gpt-5.4 / gpt-5.5 / gpt-5.6 families. The raw OpenAI API and OpenRouter
+# expose 1.05M for the same slugs, but the Codex backend hard-caps at 272K
+# (verified live for 5.4/5.5: a ~330K-token request to
 # chatgpt.com/backend-api/codex/responses is rejected with
-# ``context_length_exceeded`` while ~250K succeeds). With a 272K ceiling the
-# default 50% compaction trigger fires at ~136K — wasteful, since the model
-# can hold far more raw context before summarization actually buys anything.
-# We raise the trigger to 85% (~231K) on this exact route so Codex gpt-5.4/
-# gpt-5.5 sessions use the window they actually have.
+# ``context_length_exceeded`` while ~250K succeeds; gpt-5.6 shares the same
+# 272K Codex cap — see _CODEX_OAUTH_CONTEXT_FALLBACK in model_metadata.py).
+# With a 272K ceiling the default 50% compaction trigger fires at ~136K —
+# wasteful, since the model can hold far more raw context before
+# summarization actually buys anything. We raise the trigger to 85% (~231K)
+# on this exact route so Codex gpt-5.4 / gpt-5.5 / gpt-5.6 sessions use the
+# window they actually have.
 _CODEX_GPT54_GPT55_COMPACTION_THRESHOLD = 0.85
 
 # gpt-5.3-codex-spark is Codex-OAuth-only (ChatGPT Pro entitlement) with a
@@ -336,14 +339,16 @@ _CODEX_SPARK_COMPACTION_THRESHOLD = 0.70
 
 
 def _is_codex_gpt54_or_gpt55(model: Optional[str], provider: Optional[str] = None) -> bool:
-    """True for gpt-5.4 / gpt-5.5 on the ChatGPT Codex OAuth backend.
+    """True for gpt-5.4 / gpt-5.5 / gpt-5.6 on the ChatGPT Codex OAuth backend.
 
     Matches only the Codex OAuth route (provider ``openai-codex``), not the
     direct OpenAI API, OpenRouter, or GitHub Copilot paths — those expose a
     larger context window for the same slug and must keep the user's default
-    compaction threshold. ``gpt-5.4-pro`` / ``gpt-5.5-pro`` and dated snapshots
-    are matched via prefix so the override tracks both 272K-capped families
-    without re-listing every variant.
+    compaction threshold. ``-pro`` variants and dated snapshots are matched
+    via prefix so the override tracks every 272K-capped family (5.4, 5.5,
+    5.6 sol/terra/luna incl. their ``-pro`` modes) without re-listing every
+    variant. (Name kept for backward compatibility with the
+    ``compression.codex_gpt55_autoraise`` config key.)
     """
     prov = (provider or "").strip().lower()
     if prov != "openai-codex":
@@ -356,6 +361,9 @@ def _is_codex_gpt54_or_gpt55(model: Optional[str], provider: Optional[str] = Non
         or bare == "gpt-5.5"
         or bare.startswith("gpt-5.5-")
         or bare.startswith("gpt-5.5.")
+        or bare == "gpt-5.6"
+        or bare.startswith("gpt-5.6-")
+        or bare.startswith("gpt-5.6.")
     )
 
 
@@ -410,11 +418,12 @@ def _compression_threshold_for_model(
 
     Per-model/route overrides:
       - Arcee Trinity Large Thinking → 0.75 (preserve reasoning context).
-      - gpt-5.4 / gpt-5.5 on the Codex OAuth route → 0.85, because Codex caps
-        both families at 272K and the default 50% trigger would compact at
-        ~136K. Gated by ``allow_codex_gpt55_autoraise`` (historical config-key
-        name kept for backward compatibility) so the user can opt back down to
-        the global default (the caller passes the config flag through here).
+      - gpt-5.4 / gpt-5.5 / gpt-5.6 on the Codex OAuth route → 0.85, because
+        Codex caps all three families at 272K and the default 50% trigger
+        would compact at ~136K. Gated by ``allow_codex_gpt55_autoraise``
+        (historical config-key name kept for backward compatibility) so the
+        user can opt back down to the global default (the caller passes the
+        config flag through here).
       - gpt-5.3-codex-spark on the Codex OAuth route → 0.70, because the model
         has a native 128K window and the default 50% trigger would compact at
         ~64K — wasting half the usable context. Not gated by the gpt-5.5
diff --git a/hermes_cli/codex_models.py b/hermes_cli/codex_models.py
index bd7b2d78e8b..a56cddf73a2 100644
--- a/hermes_cli/codex_models.py
+++ b/hermes_cli/codex_models.py
@@ -12,10 +12,14 @@ import os
 logger = logging.getLogger(__name__)
 
 DEFAULT_CODEX_MODELS: List[str] = [
-    # GPT-5.6 series (Sol/Terra/Luna) — GA 2026-07-09 (previewed 2026-06-26).
+    # GPT-5.6 series (Sol/Terra/Luna + -pro high-effort modes) — GA 2026-07-09
+    # (previewed 2026-06-26).
     "gpt-5.6-sol",
+    "gpt-5.6-sol-pro",
     "gpt-5.6-terra",
+    "gpt-5.6-terra-pro",
     "gpt-5.6-luna",
+    "gpt-5.6-luna-pro",
     "gpt-5.5",
     "gpt-5.4-mini",
     "gpt-5.4",
@@ -49,8 +53,11 @@ DEFAULT_CODEX_MODELS: List[str] = [
 
 _FORWARD_COMPAT_TEMPLATE_MODELS: List[tuple[str, tuple[str, ...]]] = [
     ("gpt-5.6-sol", ("gpt-5.5", "gpt-5.4")),
+    ("gpt-5.6-sol-pro", ("gpt-5.5", "gpt-5.4")),
     ("gpt-5.6-terra", ("gpt-5.5", "gpt-5.4")),
+    ("gpt-5.6-terra-pro", ("gpt-5.5", "gpt-5.4")),
     ("gpt-5.6-luna", ("gpt-5.5", "gpt-5.4")),
+    ("gpt-5.6-luna-pro", ("gpt-5.5", "gpt-5.4")),
     ("gpt-5.5", ("gpt-5.4", "gpt-5.4-mini", "gpt-5.3-codex")),
     ("gpt-5.4-mini", ("gpt-5.3-codex",)),
     ("gpt-5.4", ("gpt-5.3-codex",)),
diff --git a/hermes_cli/config.py b/hermes_cli/config.py
index 7c83c6f167a..678cdb869af 100644
--- a/hermes_cli/config.py
+++ b/hermes_cli/config.py
@@ -1423,17 +1423,18 @@ DEFAULT_CONFIG = {
                                       # True if you'd rather pause than silently lose
                                       # context turns when your aux model is flaky.
         "codex_gpt55_autoraise": True,  # Historical key name kept for compatibility.
-                                      # When True, gpt-5.4 / gpt-5.5 on the ChatGPT Codex
-                                      # OAuth route raise their compaction trigger to 85%
-                                      # (vs the global `threshold` above). Codex hard-caps
-                                      # both families at a 272K window, so the default 50%
-                                      # would compact at ~136K and waste half the usable
-                                      # context. Set to False to opt back down to the global
-                                      # threshold (e.g. 0.50) for those Codex sessions.
-                                      # Only this exact route is affected — gpt-5.4 / 5.5
-                                      # on OpenAI's direct API, OpenRouter, and Copilot keep
-                                      # the global threshold regardless.
-        "codex_gpt55_autoraise_notice": True,  # Display the one-time Codex gpt-5.4/5.5
+                                      # When True, gpt-5.4 / gpt-5.5 / gpt-5.6 on the
+                                      # ChatGPT Codex OAuth route raise their compaction
+                                      # trigger to 85% (vs the global `threshold` above).
+                                      # Codex hard-caps these families at a 272K window, so
+                                      # the default 50% would compact at ~136K and waste half
+                                      # the usable context. Set to False to opt back down to
+                                      # the global threshold (e.g. 0.50) for those Codex
+                                      # sessions. Only this exact route is affected —
+                                      # gpt-5.4 / 5.5 / 5.6 on OpenAI's direct API,
+                                      # OpenRouter, and Copilot keep the global threshold
+                                      # regardless.
+        "codex_gpt55_autoraise_notice": True,  # Display the one-time Codex gpt-5.4/5.5/5.6
                                       # autoraise banner. Set False to keep the
                                       # 85% threshold autoraise but suppress the
                                       # user-facing notice in CLI/gateway output.
diff --git a/tests/hermes_cli/test_gpt56_registration.py b/tests/hermes_cli/test_gpt56_registration.py
index 9d63992ecaa..5799fa08358 100644
--- a/tests/hermes_cli/test_gpt56_registration.py
+++ b/tests/hermes_cli/test_gpt56_registration.py
@@ -81,3 +81,53 @@ class TestGpt56PricingRoute:
                 _OFFICIAL_DOCS_PRICING[("openai", f"{base}-pro")]
                 is _OFFICIAL_DOCS_PRICING[("openai", base)]
             ), base
+
+
+class TestGpt56CodexCompaction:
+    """Codex OAuth caps the whole gpt-5.6 family at 272K, same as 5.4/5.5, so
+    the compaction auto-raise (0.85) must fire for every 5.6 variant on the
+    openai-codex route and NOT on the direct-API/OpenRouter routes."""
+
+    def test_autoraise_applies_to_all_56_on_codex(self):
+        from agent.auxiliary_client import _compression_threshold_for_model
+
+        for slug in (
+            "gpt-5.6-sol",
+            "gpt-5.6-sol-pro",
+            "gpt-5.6-terra",
+            "gpt-5.6-terra-pro",
+            "gpt-5.6-luna",
+            "gpt-5.6-luna-pro",
+        ):
+            assert (
+                _compression_threshold_for_model(slug, provider="openai-codex")
+                == 0.85
+            ), slug
+
+    def test_no_autoraise_on_direct_api_route(self):
+        from agent.auxiliary_client import _compression_threshold_for_model
+
+        # Direct OpenAI API / OpenRouter expose the full 1.05M window, so the
+        # 272K-cap override must NOT apply there.
+        assert (
+            _compression_threshold_for_model("gpt-5.6-sol", provider="openai")
+            is None
+        )
+        assert (
+            _compression_threshold_for_model(
+                "openai/gpt-5.6-sol", provider="openrouter"
+            )
+            is None
+        )
+
+    def test_autoraise_respects_opt_out(self):
+        from agent.auxiliary_client import _compression_threshold_for_model
+
+        assert (
+            _compression_threshold_for_model(
+                "gpt-5.6-sol",
+                provider="openai-codex",
+                allow_codex_gpt55_autoraise=False,
+            )
+            is None
+        )