run llama-3.1-8b on 3 m4 pro cluster

.github/workflows/bench_job.yml

@@ -80,6 +80,9 @@ jobs:
           
           # Source the performance environment variables
           source /tmp/performance_env.sh
+
+          # MLX Memory Settings
+          ./configure_mlx.sh
           
           # Verify optimizations
           echo "Verifying performance settings..."

.github/workflows/benchmarks.yml

@@ -24,7 +24,7 @@ jobs:
   three-m4-pro-cluster:
     strategy:
       matrix:
-        model: ['llama-3.2-1b', 'llama-3.2-3b']
+        model: ['llama-3.2-1b', 'llama-3.2-3b', 'llama-3.1-8b']
       # Optional: add fail-fast: false if you want all matrix jobs to continue even if one fails
       fail-fast: false
     uses: ./.github/workflows/bench_job.yml