shorter

.github/bench.py

@@ -19,7 +19,7 @@ async def measure_performance(api_endpoint: str, prompt: str) -> Dict[str, Any]:
     """
     results: Dict[str, Any] = {}
     request_payload = {
-        "model": "llama-3.2-3b",
+        "model": "llama-3.2-1b",
         "messages": [{"role": "user", "content": prompt}],
         "temperature": 0,
         "stream": True

.github/workflows/build_and_test.yml

@@ -46,7 +46,7 @@ jobs:
           source env/bin/activate
           export PATH="/usr/local/bin:/opt/homebrew/bin:$PATH"
           echo "${ALL_NODE_IDS}"
-          DEBUG_DISCOVERY=7 DEBUG=7 exo --node-id="${MY_NODE_ID}" --node-id-filter="${ALL_NODE_IDS}" --chatgpt-api-port 52415 --disable-tui > output1.log 2>&1 &
+          DEBUG_DISCOVERY=7 DEBUG=7 exo --max-generate-tokens=100 --node-id="${MY_NODE_ID}" --node-id-filter="${ALL_NODE_IDS}" --chatgpt-api-port 52415 --disable-tui > output1.log 2>&1 &
           PID1=$!
           tail -f output1.log &
           TAIL1=$!