fast-inference-nllb

Paused

fast-inference-nllb / Dockerfile

Update Dockerfile

d17e482 verified 2 months ago

189 Bytes

	FROM ghcr.io/winstxnhdw/nllb-api:main

	ENV SERVER_PORT=7860
	ENV OMP_NUM_THREADS=4
	ENV CT2_USE_EXPERIMENTAL_PACKED_GEMM=1
	ENV CT2_FORCE_CPU_ISA=AVX512
	ENV WORKER_COUNT=2

	EXPOSE $SERVER_PORT