jandres/GenAIExamples
1
0
Fork 0
Code Issues Pull Requests Actions 5 Packages Projects Releases Wiki Activity

Use official tei gaudi image and update tgi gaudi version (#810)

Browse Source 
Signed-off-by: lvliang-intel <liang1.lv@intel.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
This commit is contained in:
lvliang-intel
2024-09-23 17:52:56 +08:00
committed by GitHub
parent c35fe0b429
commit 3fb60608b3
72 changed files with 8024 additions and 154 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
Translation/docker_compose/intel/hpu/gaudi/compose.yaml
View File

@@ -3,18 +3,23 @@
services:
tgi-service:
image: ghcr.io/huggingface/tgi-gaudi:2.0.1
image: ghcr.io/huggingface/tgi-gaudi:2.0.5
container_name: tgi-gaudi-server
ports:
- "8008:80"
environment:
no_proxy: ${no_proxy}
http_proxy: ${http_proxy}
https_proxy: ${https_proxy}
HF_TOKEN: ${HUGGINGFACEHUB_API_TOKEN}
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
ENABLE_HPU_GRAPH: true
LIMIT_HPU_GRAPH: true
USE_FLASH_ATTENTION: true
FLASH_ATTENTION_RECOMPUTE: true
runtime: habana
cap_add:
- SYS_NICE
@@ -31,6 +36,7 @@ services:
- "9000:9000"
ipc: host
environment:
no_proxy: ${no_proxy}
http_proxy: ${http_proxy}
https_proxy: ${https_proxy}
TGI_LLM_ENDPOINT: ${TGI_LLM_ENDPOINT}
@@ -47,6 +53,7 @@ services:
ports:
- "8888:8888"
environment:
- no_proxy=${no_proxy}
- https_proxy=${https_proxy}
- http_proxy=${http_proxy}
- MEGA_SERVICE_HOST_IP=${MEGA_SERVICE_HOST_IP}
@@ -61,6 +68,7 @@ services:
ports:
- "5173:5173"
environment:
- no_proxy=${no_proxy}
- https_proxy=${https_proxy}
- http_proxy=${http_proxy}
- BASE_URL=${BACKEND_SERVICE_ENDPOINT}