Skip to content

Commit f40d7c6

Browse files
yyssysyyssys
authored
[Docs]When XPU starts the service, the model loader uses the default version (#4292)
1 parent 331c4d2 commit f40d7c6

File tree

2 files changed

+32
-30
lines changed

2 files changed

+32
-30
lines changed

‎docs/usage/kunlunxin_xpu_deployment.md‎

Lines changed: 16 additions & 15 deletions
Original file line numberDiff line numberDiff line change
@@ -1,20 +1,20 @@
11
## Supported Models
22
|Model Name|Context Length|Quantization|XPUs Required|Deployment Commands|Minimum Version Required|
33
|-|-|-|-|-|-|
4-
|ERNIE-4.5-300B-A47B|32K|WINT8|8|export XPU_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-300B-A47B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 8 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 64 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9|>=2.0.3|
5-
|ERNIE-4.5-300B-A47B|32K|WINT4|4 (Recommended)|export XPU_VISIBLE_DEVICES="0,1,2,3" or "4,5,6,7"<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-300B-A47B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 4 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 64 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.9|>=2.0.0|
6-
|ERNIE-4.5-300B-A47B|32K|WINT4|8|export XPU_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-300B-A47B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 8 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 64 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.95|>=2.0.0|
7-
|ERNIE-4.5-300B-A47B|128K|WINT4|8 (Recommended)|export XPU_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-300B-A47B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 8 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 64 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.9|>=2.0.0|
8-
|ERNIE-4.5-21B-A3B|32K|BF16|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --gpu-memory-utilization 0.9|>=2.1.0|
9-
|ERNIE-4.5-21B-A3B|32K|WINT8|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9|>=2.1.0|
10-
|ERNIE-4.5-21B-A3B|32K|WINT4|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.9|>=2.1.0|
11-
|ERNIE-4.5-21B-A3B|128K|BF16|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --gpu-memory-utilization 0.9|>=2.1.0|
12-
|ERNIE-4.5-21B-A3B|128K|WINT8|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9|>=2.1.0|
13-
|ERNIE-4.5-21B-A3B|128K|WINT4|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.9|>=2.1.0|
14-
|ERNIE-4.5-0.3B|32K|BF16|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-0.3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --gpu-memory-utilization 0.9|>=2.0.3|
15-
|ERNIE-4.5-0.3B|32K|WINT8|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-0.3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9|>=2.0.3|
16-
|ERNIE-4.5-0.3B|128K|BF16|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-0.3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --gpu-memory-utilization 0.9|>=2.0.3|
17-
|ERNIE-4.5-0.3B|128K|WINT8|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-0.3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9|>=2.0.3|
4+
|ERNIE-4.5-300B-A47B|32K|WINT8|8|export XPU_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-300B-A47B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 8 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 64 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.0.3|
5+
|ERNIE-4.5-300B-A47B|32K|WINT4|4 (Recommended)|export XPU_VISIBLE_DEVICES="0,1,2,3" or "4,5,6,7"<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-300B-A47B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 4 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 64 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.0.0|
6+
|ERNIE-4.5-300B-A47B|32K|WINT4|8|export XPU_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-300B-A47B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 8 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 64 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.95 \ <br> --load-choices "default"|>=2.0.0|
7+
|ERNIE-4.5-300B-A47B|128K|WINT4|8 (Recommended)|export XPU_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-300B-A47B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 8 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 64 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.0.0|
8+
|ERNIE-4.5-21B-A3B|32K|BF16|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.1.0|
9+
|ERNIE-4.5-21B-A3B|32K|WINT8|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.1.0|
10+
|ERNIE-4.5-21B-A3B|32K|WINT4|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.1.0|
11+
|ERNIE-4.5-21B-A3B|128K|BF16|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.1.0|
12+
|ERNIE-4.5-21B-A3B|128K|WINT8|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.1.0|
13+
|ERNIE-4.5-21B-A3B|128K|WINT4|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-21B-A3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --quantization "wint4" \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.1.0|
14+
|ERNIE-4.5-0.3B|32K|BF16|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-0.3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.0.3|
15+
|ERNIE-4.5-0.3B|32K|WINT8|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-0.3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 32768 \ <br> --max-num-seqs 128 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.0.3|
16+
|ERNIE-4.5-0.3B|128K|BF16|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-0.3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.0.3|
17+
|ERNIE-4.5-0.3B|128K|WINT8|1|export XPU_VISIBLE_DEVICES="0" # Specify any card<br>export ENABLE_V1_KVCACHE_SCHEDULER=0 # V1 is not supported<br>python -m fastdeploy.entrypoints.openai.api_server \ <br> --model PaddlePaddle/ERNIE-4.5-0.3B-Paddle \ <br> --port 8188 \ <br> --tensor-parallel-size 1 \ <br> --max-model-len 131072 \ <br> --max-num-seqs 128 \ <br> --quantization "wint8" \ <br> --gpu-memory-utilization 0.9 \ <br> --load-choices "default"|>=2.0.3|
1818

1919
## Quick start
2020

@@ -36,7 +36,8 @@ python -m fastdeploy.entrypoints.openai.api_server \
3636
--max-model-len 32768 \
3737
--max-num-seqs 64 \
3838
--quantization "wint4" \
39-
--gpu-memory-utilization 0.9
39+
--gpu-memory-utilization 0.9 \
40+
--load-choices "default"
4041
```
4142

4243
**Note:** When deploying on 4 XPUs, only two configurations are supported which constrained by hardware limitations such as interconnect capabilities.

0 commit comments

Comments
 (0)