Download inference-cache-config/phi4.json from aws-neuron/optimum-neuron-cache: direct link, hf CLI and curl.
- Browser
- Download file 556 Bytes
-
https://huggingface.co/aws-neuron/optimum-neuron-cache/resolve/e00ffa4c39b354aba0b67e8a2d03edb0c3ff4ecc/inference-cache-config/phi4.json
- Command line
-
hf download hf://aws-neuron/optimum-neuron-cache@e00ffa4c39b354aba0b67e8a2d03edb0c3ff4ecc/inference-cache-config/phi4.json
-
curl -L -o phi4.json https://huggingface.co/aws-neuron/optimum-neuron-cache/resolve/e00ffa4c39b354aba0b67e8a2d03edb0c3ff4ecc/inference-cache-config/phi4.json
556 Bytes
| { | |
| "microsoft/Phi-3-mini-4k-instruct": [ | |
| { | |
| "batch_size": 1, | |
| "sequence_length": 4096, | |
| "num_cores": 2, | |
| "auto_cast_type": "bf16" | |
| }, | |
| { | |
| "batch_size": 4, | |
| "sequence_length": 4096, | |
| "num_cores": 2, | |
| "auto_cast_type": "bf16" | |
| } | |
| ], | |
| "microsoft/phi-4": [ | |
| { | |
| "batch_size": 1, | |
| "sequence_length": 4096, | |
| "num_cores": 10, | |
| "auto_cast_type": "bf16" | |
| }, | |
| { | |
| "batch_size": 16, | |
| "sequence_length": 4096, | |
| "num_cores": 10, | |
| "auto_cast_type": "bf16" | |
| } | |
| ] | |
| } |