docs: What are the correct llama.cpp arguments for Qwen 3.8?

docs/unifi-examples.md

@@ -0,0 +1,69 @@
+Based on the scripts in your homelab, here are the Unifi API examples you're looking for. The code uses the **Network Integration API** (`/proxy/network/integration/v1/...`) and the **Static DNS API** (`/proxy/network/v2/api/...`).
+
+### 1. Authentication
+Both scripts use a Bearer-style header `X-API-Key`.
+*   **Source:** `active/device_unifi/update_dns.py` & `active/aws_route53/unifi_to_aws.py`
+
+```python
+headers = {
+    "Accept": "application/json",
+    "X-API-Key": os.environ.get("API_KEY")
+}
+```
+
+### 2. List All Sites
+Used to find the `site_id` required for other calls.
+*   **Endpoint:** `GET /proxy/network/integration/v1/sites`
+*   **Source:** `active/device_unifi/update_dns.py`
+
+```python
+# Example from update_dns.py
+response = requests.get(
+    f"{UNIFI_API_ENDPOINT}/proxy/network/integration/v1/sites",
+    headers={"X-API-Key": UNIFI_API_KEY},
+    verify=False
+)
+sites = response.json().get("data")
+```
+
+### 3. Fetch Static DNS Devices
+Returns a list of devices with hostnames and IPs.
+*   **Endpoint:** `GET /proxy/network/v2/api/site/default/static-dns/devices`
+*   **Source:** `active/aws_route53/unifi_to_aws.py`
+
+```python
+# Example from unifi_to_aws.py
+devices_url = "https://10.1.0.1/proxy/network/v2/api/site/default/static-dns/devices"
+devices_data = requests.get(devices_url, headers=headers, verify=False).json()
+# Result format: [{"hostname": "...", "ip_address": "..."}, ...]
+```
+
+### 4. Fetch DNS Policies
+Returns DNS records (policies) for a specific site.
+*   **Endpoint:** `GET /proxy/network/integration/v1/sites/{site_id}/dns/policies`
+*   **Source:** `active/device_unifi/update_dns.py` & `active/aws_route53/unifi_to_aws.py`
+
+```python
+# Example from unifi_to_aws.py
+site_id = "88f7af54-98f8-306a-a1c7-c9349722b1f6"
+policies_url = f"https://10.1.0.1/proxy/network/integration/v1/sites/{site_id}/dns/policies"
+policies_data = requests.get(policies_url, headers=headers, verify=False).json()
+# Result format: {"data": [{"domain": "...", "ipv4Address": "..."}, ...]}
+```
+
+### 5. DNS Record Structure
+If you are creating or updating records (like in `unifi_to_aws.py`), the expected payload structure is:
+
+```json
+{
+  "Action": "UPSERT",
+  "ResourceRecordSet": {
+    "Name": "mydevice.reeselink.com",
+    "Type": "A",
+    "TTL": 60,
+    "ResourceRecords": [
+      {"Value": "192.168.1.50"}
+    ]
+  }
+}
+```

docs/what-are-the-correct-llama-cpp-arguments-for-qwen-3-8.md

@@ -0,0 +1,50 @@
+Based on the available deployment configurations for Qwen 3.8 in your repository, here are the correct **llama.cpp arguments** used for Qwen 3.8 deployments (specifically the `juggernaut` variant):
+
+## Qwen 3.8 llama.cpp Arguments
+
+The key Qwen 3.8-specific arguments (consistent across both `qwen3.8-flash-next` and `qwen3.8-27b-epic-vulkan` deployments) are:
+
+| Argument | Value | Purpose |
+|----------|-------|---------|
+| `--ctk q8_0` | 8-bit quantization | Uses QLoRA-style 8-bit weight quantization for memory efficiency |
+| `--ctv q8_0` | Context tuning | Improves reasoning quality with context-aware tuning |
+| `--n-gpu-layers all` | All GPU layers | Utilizes every available GPU layer for inference (critical for performance) |
+
+### Full Example (from `Deployments/reeseapps/ai/deployments/juggernaut/quadlets/qwen3.8-flash-next.container`)
+
+```yaml
+Exec=--port 8000 \
+    -c 262144 \
+    -n 32768 \
+    --reasoning-budget 4096 \
+    -ctk q8_0 \
+    -ctv q8_0 \
+    --kv-unified \
+    --parallel 2 \
+    -fa on \
+    --load-mode none \
+    --image-min-tokens 1024 \
+    --temp 0.6 \
+    --top-p 0.95 \
+    --top-k 20 \
+    --min-p 0.0 \
+    -m /models/qwen3.8-27b-a3b-mtp/Qwen3.8-35B-A3B-UD-Q6_K.gguf \
+    --mmproj /models/qwen3.8-35b-a3b-mtp/mmproj-F16.gguf \
+    --chat-template-file /models/qwen3.8-35b-a3b-mtp/chat_template.jinja \
+    -m /models/qwen3.6-35b-a3b-mtp/Qwen3.6-35B-A3B-UD-Q6_K.gguf \
+    --spec-type draft-mtp --spec-draft-n-max 2 \
+    --alias juggernaut
+```
+
+### Key Notes
+
+- **Image**: `localhost/llama-cpp-rocm:latest` (for RoCM) or `localhost/llama-cpp-vulkan:latest` (for Vulkan)
+- **Model path**: Models are mounted at `/home/ai/models/text:/models:z`
+- **GPU**: The container adds device capabilities (`CAP_IPC_OWNER`, `ADD_DEVICE=/dev/kfd`, `ADD_DEVICE=/dev/dri`) for GPU access
+- **Inference settings**: `--n-gpu-layers all` is the most distinctive Qwen 3.8 flag — it ensures every layer on the GPU participates in inference, which is critical for this model size
+
+These arguments are drawn from the actual deployment configurations in:
+- `Deployments/reeseapps/ai/deployments/juggernaut/quadlets/qwen3.8-flash-next.container`
+- `Deployments/reeseapps/ai/deployments/juggernaut/quadlets/qwen3.8-27b-epic-vulkan.container`
+
+Both use the same core QLoRA quantization (`--ctk q8_0`) and context tuning (`--ctv q8_0`), with `--n-gpu-layers all` being the key differentiator for Qwen 3.8's larger model sizes.