feat(litellm): add mistral-medium-3.5 (RecViking NVFP4 :8012, temporary GPU0 tenant)

stacks/litellm/conf/config.yaml

@@ -95,6 +95,19 @@ model_list:
     model_info:
       mode: chat
 
+  # --- Mistral Medium 3.5 (RecViking NVFP4, vLLM nightly) — TEMPORARY speed-check
+  #     tenant DISPLACING mistral-small-4 on GPU 0 (2026-06-19). Loads the HF-format
+  #     NVFP4 directly on :8012; served-name mistral-medium-3.5. NOTE: while this is
+  #     up, mistral-small-4 (above) points at a DOWN backend — revert is the
+  #     mistral-small-4-heretic stack. ---
+  - model_name: mistral-medium-3.5
+    litellm_params:
+      model: hosted_vllm/mistral-medium-3.5
+      api_base: http://10.250.50.54:8012/v1
+      api_key: os.environ/VLLM_API_KEY
+    model_info:
+      mode: chat
+
   # --- Selene 1 Mini 8B (AtlaAI judge, FP8) — restored on GPU1 after the
   #     llama-swap teardown (was the Q6_K GGUF in the swap zoo). vLLM dynamic fp8,
   #     :8011. Explicit entry shadows the "*" wildcard (which used to reach it via