diff --git a/README.md b/README.md
index 11e7de0..8b8f9cb 100644
--- a/README.md
+++ b/README.md
@@ -6,6 +6,8 @@
 
 SentinelOps ingests real-time logs from distributed microservices, detects anomalies using statistical models (z-score + EWMA), and will expose a natural-language incident-query interface (LLM/RAG) for fast triage. Built as a polyglot system (Go, Rust, Python) to demonstrate production patterns across the stack: event-driven ingestion, self-migrating services, batched writes, full observability, and a real CI/CD pipeline with branch protection.
 
+![SentinelOps demo](docs/media/demo.gif)
+
 ## What's actually running today
 
 ```mermaid
diff --git a/docs/media/demo.gif b/docs/media/demo.gif
new file mode 100644
index 0000000..250863f
Binary files /dev/null and b/docs/media/demo.gif differ
diff --git a/infra/docker-compose.yml b/infra/docker-compose.yml
index 3df55cb..107cb8f 100644
--- a/infra/docker-compose.yml
+++ b/infra/docker-compose.yml
@@ -134,10 +134,10 @@ services:
     environment:
       DATABASE_URL: postgres://sentinelops:${TIMESCALE_PASSWORD:-devpassword}@timescaledb:5432/sentinelops
       OLLAMA_BASE_URL: http://host.docker.internal:11434
-      OLLAMA_MODEL: llama3.2:3b
+      OLLAMA_MODEL: llama3.2:1b
       OLLAMA_KEEP_ALIVE: 10m
-      OLLAMA_NUM_PREDICT: "300"
-      OLLAMA_TIMEOUT_SECONDS: "180"
+      OLLAMA_NUM_PREDICT: "120"
+      OLLAMA_TIMEOUT_SECONDS: "240"
     extra_hosts:
       - "host.docker.internal:host-gateway"
     ports:
diff --git a/services/llm-query-python/Dockerfile b/services/llm-query-python/Dockerfile
index 4376520..77d3151 100644
--- a/services/llm-query-python/Dockerfile
+++ b/services/llm-query-python/Dockerfile
@@ -9,7 +9,7 @@ COPY app/ ./app/
 
 ENV DATABASE_URL=postgresql://sentinelops:devpassword@timescaledb:5432/sentinelops \
     OLLAMA_BASE_URL=http://host.docker.internal:11434 \
-    OLLAMA_MODEL=llama3.2:3b \
+    OLLAMA_MODEL=llama3.2:1b \
     METRICS_PORT=9300
 
 EXPOSE 9300
diff --git a/services/llm-query-python/app/config.py b/services/llm-query-python/app/config.py
index 2c91b3c..00cc4ce 100644
--- a/services/llm-query-python/app/config.py
+++ b/services/llm-query-python/app/config.py
@@ -8,10 +8,10 @@ class Settings(BaseSettings):
         "postgresql://sentinelops:devpassword@localhost:5432/sentinelops"
     )
     ollama_base_url: str = "http://host.docker.internal:11434"
-    ollama_model: str = "llama3.2:3b"
+    ollama_model: str = "llama3.2:1b"
     ollama_keep_alive: str = "10m"
-    ollama_num_predict: int = 300
-    ollama_timeout_seconds: float = 180.0
+    ollama_num_predict: int = 120
+    ollama_timeout_seconds: float = 240.0
     default_lookback_hours: int = 6
     metrics_port: int = 9300