gemma-4-E2B-coreml / prefill /chunk2.mlmodelc

Commit History

prefill: bump prefillN from 64 to 512 (~8x faster TTFT on 50-500 token prompts)
dfaa01b
verified

mlboydaisuke commited on

Upload prefill/chunk2.mlmodelc/analytics/coremldata.bin with huggingface_hub
9f95465
verified

mlboydaisuke commited on

Upload prefill/chunk2.mlmodelc/weights/weight.bin with huggingface_hub
57c97a3
verified

mlboydaisuke commited on

Upload prefill/chunk2.mlmodelc/coremldata.bin with huggingface_hub
f528c21
verified

mlboydaisuke commited on

Upload prefill/chunk2.mlmodelc/model.mil with huggingface_hub
1e10260
verified

mlboydaisuke commited on