Default to DeepSeek V4 Pro (best-in-class agentic, uncensored via ghost's steer)

README.md

@@ -33,7 +33,7 @@ Then connect your account once and run **`ghost`**:
 
 ```bash
 ghost-login        # browser login -> hands a session token back to this machine
-ghost              # chat (default = Hermes 405B via the OpenGradient TEE gateway, OHTTP-private)
+ghost              # chat (default = DeepSeek V4 Pro via the OpenGradient TEE gateway, OHTTP-private)
 ghost --local      # force the fully-offline local model (no scrubber, no og-veil, nothing leaves)
 ```
 
@@ -99,7 +99,7 @@ The default is the hosted Hermes 405B because it is the stronger agent; the OHTT
 
 | Layer | Behaviour |
 |---|---|
-| **Default model** | `nous/hermes-4-405b` via the `opengradient` provider -- scrubber (`:8788`) → og-veil (`:11435`) → chat-api relay → TEE gateway |
+| **Default model** | `deepseek/deepseek-v4-pro` via the `opengradient` provider -- scrubber (`:8788`) → og-veil (`:11435`) → chat-api relay → TEE gateway (uncensored through ghost's steer; strongest open agentic model). `nous/hermes-4-405b` and the rest of the catalog via `/model`. |
 | **Hosted line-up** | Hermes 4 (405B / 70B) -- unrestricted, open-weight models only, over the one OHTTP path. Closed/refusing models (Claude, GPT, Gemini, Grok) are deliberately not offered. |
 | **Fallback model** | `fallback_model` → local `uncensored-local` (32B) if the hosted gateway is unreachable |
 | **Tool / auxiliary model** | Local 7B abliterated (`ghost-tool`) runs titling, compression, triage -- never a hosted provider |
@@ -228,7 +228,7 @@ ghost-login --status        # who am I logged in as?
 ```
 
 ```bash
-ghost                       # chat (default = Hermes 405B via the TEE gateway, OHTTP-private)
+ghost                       # chat (default = DeepSeek V4 Pro via the TEE gateway, OHTTP-private)
 ghost --yolo -z "..."       # one-shot
 ghost --paths "..."         # agentic file work: real filesystem paths reach the hosted model
                             #   (your name + secrets in content are still scrubbed)

bin/ghost

@@ -78,9 +78,9 @@ if [ "$HC" = "200" ] && [ "$VC" = "200" ]; then
   # IP-masked through it -- we probe its exit IP purely for the status line.
   EXIT_IP=$(/usr/bin/curl -s --max-time 6 -x "$PROXY" https://api.ipify.org 2>/dev/null)
   if [ -n "$EXIT_IP" ]; then
-    STATUS="👻 ghost · default = Hermes 405B via TEE gateway (OHTTP/og-veil) · IP-masked via $EXIT_IP · $LOGIN · $FB"
+    STATUS="👻 ghost · default = DeepSeek V4 Pro via TEE gateway (OHTTP/og-veil) · IP-masked via $EXIT_IP · $LOGIN · $FB"
   else
-    STATUS="👻 ghost · direct · default = Hermes 405B via TEE gateway (OHTTP/og-veil) · $LOGIN · $FB"
+    STATUS="👻 ghost · direct · default = DeepSeek V4 Pro via TEE gateway (OHTTP/og-veil) · $LOGIN · $FB"
   fi
 elif [ "$HC" = "200" ]; then
   STATUS="👻 ghost · ⚠️ og-veil DOWN (veil=$VC) -- hosted unreachable (try ghost-login), $FB_DOWN"

privacy/scrubbing_proxy.py

@@ -71,8 +71,8 @@
 # gateway serves them. Do NOT trust that list to add a model; probe the endpoint. The gateway
 # rejects anything it doesn't serve with `Model '<id>' is not supported`.
 _CATALOG_MODELS = [
-    ("nous/hermes-4-405b", "Hermes 4 405B — flagship uncensored open model, most steerable (default)"),
-    ("deepseek/deepseek-v4-pro", "DeepSeek V4 Pro — strongest open reasoning + coding; best for agentic work"),
+    ("deepseek/deepseek-v4-pro", "DeepSeek V4 Pro — strongest open reasoning + coding; best for agentic work (default)"),
+    ("nous/hermes-4-405b", "Hermes 4 405B — flagship uncensored open model, most steerable"),
     ("zai/glm-5.2", "GLM 5.2 — strong open agentic MoE (Z.ai)"),
     ("nous/hermes-4-70b", "Hermes 4 70B — fast, low-cost open-weight model"),
 ]

profile/config.yaml

@@ -1,6 +1,6 @@
 model:
   base_url: http://127.0.0.1:8788/v1
-  default: nous/hermes-4-405b
+  default: deepseek/deepseek-v4-pro
   provider: opengradient
   context_length: 65536
   ollama_num_ctx: 65536
@@ -17,7 +17,7 @@ providers:
   opengradient:
     base_url: http://127.0.0.1:8788/v1
     api_key: managed-by-og-veil
-    default_model: nous/hermes-4-405b
+    default_model: deepseek/deepseek-v4-pro
     context_length: 65536
   ollama-local:
     base_url: http://localhost:11434/v1