- Llama3_1_8B:
- Multiple Steps:
- GPU Memory Usage: 6098 MiB
- Total CPU Usage: 43.2% (6.912 cores)
- Total Memory Usage: 3% (3,932.16 MiB)
- Long Context:
- GPU Memory Usage: 6132 MiB
- Total CPU Usage: 37% (5.92 cores)
- Total Memory Usage: 3% (3,932.16 MiB)
- Phi3Mini:
- Multiple Steps:
- Memory Usage: 6132 MiB
- Total CPU Usage: 147.6% (23.616 cores, can exceed 16 cores)
- Total Memory Usage: 4.6% (6,027.52 MiB)
- Long Context:
- Memory Usage: 6048 MiB
- Total CPU Usage: 70.1% (11.216 cores)
- Total Memory Usage: 2.5% (3,276.80 MiB)
- Phi3Medium128k:
- Multiple Steps:
- Memory Usage: 10956 MiB
- Total CPU Usage: 85.8% (13.728 cores)
- Total Memory Usage: 2.7% (3,531.84 MiB)
- Long Context:
- Memory Usage: 6048 MiB
- Total CPU Usage: 133.5% (21.36 cores, can exceed 16 cores)
- Total Memory Usage: 2.7% (3,531.84 MiB)
- Phi3Medium:
- Multiple Steps:
- Memory Usage: 10916 MiB
- Total CPU Usage: 136.4% (21.824 cores)
- Total Memory Usage: 2.7% (3,531.84 MiB)
- Long Context:
- Memory Usage: 10958 MiB
- Total CPU Usage: 121.1% (19.376 cores)
- Total Memory Usage: 2.7% (3,531.84 MiB)
- NousTheta:
- Multiple Steps:
- Memory Usage: 9574 MiB
- Total CPU Usage: 126.8% (20.288 cores)
- Total Memory Usage: 3.5% (4,586.8 MiB)
- Long Context:
- Memory Usage: 9582 MiB
- Total CPU Usage: 119.7% (19.152 cores)
- Total Memory Usage: 3.5% (4,586.8 MiB)