Files
GenAIExamples/AgentQnA/docker_compose/intel/hpu/gaudi/tgi_gaudi.yaml
Xiaotian Chen 1bd56af994 Update TGI image versions (#1625)
Signed-off-by: xiaotia3 <xiaotian.chen@intel.com>
2025-04-01 11:27:51 +08:00

31 lines
944 B
YAML

# Copyright (C) 2024 Intel Corporation
# SPDX-License-Identifier: Apache-2.0
services:
tgi-server:
image: ghcr.io/huggingface/tgi-gaudi:2.3.1
container_name: tgi-server
ports:
- "8085:80"
volumes:
- ${HF_CACHE_DIR}:/data
environment:
no_proxy: ${no_proxy}
http_proxy: ${http_proxy}
https_proxy: ${https_proxy}
HUGGING_FACE_HUB_TOKEN: ${HUGGINGFACEHUB_API_TOKEN}
HF_HUB_DISABLE_PROGRESS_BARS: 1
HF_HUB_ENABLE_HF_TRANSFER: 0
HABANA_VISIBLE_DEVICES: all
OMPI_MCA_btl_vader_single_copy_mechanism: none
PT_HPU_ENABLE_LAZY_COLLECTIVES: true
ENABLE_HPU_GRAPH: true
LIMIT_HPU_GRAPH: true
USE_FLASH_ATTENTION: true
FLASH_ATTENTION_RECOMPUTE: true
runtime: habana
cap_add:
- SYS_NICE
ipc: host
command: --model-id ${LLM_MODEL_ID} --max-input-length 4096 --max-total-tokens 8192 --sharded true --num-shard ${NUM_SHARDS}