-
Notifications
You must be signed in to change notification settings - Fork 37
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Llama 3.2 11B Instruct vision on 1 x L4 GPU (#258)
- Loading branch information
Showing
4 changed files
with
77 additions
and
2 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,32 @@ | ||
# Llama 3.2 11B Vision Instruct vLLM Benchmarks | ||
|
||
|
||
Single L4 GPU vLLM 0.6.2 | ||
``` | ||
python3 benchmark_serving.py --backend openai \ | ||
--base-url http://localhost:8000/openai \ | ||
--dataset-name=sharegpt --dataset-path=ShareGPT_V3_unfiltered_cleaned_split.json \ | ||
--model meta-llama-3.2-11b-vision-instruct \ | ||
--seed 12345 --tokenizer neuralmagic/Llama-3.2-11B-Vision-Instruct-FP8-dynamic | ||
============ Serving Benchmark Result ============ | ||
Successful requests: 1000 | ||
Benchmark duration (s): 681.93 | ||
Total input tokens: 230969 | ||
Total generated tokens: 194523 | ||
Request throughput (req/s): 1.47 | ||
Output token throughput (tok/s): 285.25 | ||
Total Token throughput (tok/s): 623.95 | ||
---------------Time to First Token---------------- | ||
Mean TTFT (ms): 319146.12 | ||
Median TTFT (ms): 322707.98 | ||
P99 TTFT (ms): 642512.79 | ||
-----Time per Output Token (excl. 1st token)------ | ||
Mean TPOT (ms): 54.84 | ||
Median TPOT (ms): 53.66 | ||
P99 TPOT (ms): 83.75 | ||
---------------Inter-token Latency---------------- | ||
Mean ITL (ms): 54.09 | ||
Median ITL (ms): 47.44 | ||
P99 ITL (ms): 216.77 | ||
================================================== | ||
``` |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,23 @@ | ||
# Source: models/templates/models.yaml | ||
apiVersion: kubeai.org/v1 | ||
kind: Model | ||
metadata: | ||
name: llama-3.2-11b-vision-instruct-l4 | ||
spec: | ||
features: [TextGeneration] | ||
owner: | ||
url: hf://neuralmagic/Llama-3.2-11B-Vision-Instruct-FP8-dynamic | ||
engine: VLLM | ||
args: | ||
- --max-model-len=8192 | ||
- --max-num-batched-token=8192 | ||
- --gpu-memory-utilization=0.99 | ||
- --enforce-eager | ||
- --disable-log-requests | ||
- --max-num-seqs=16 | ||
env: | ||
VLLM_WORKER_MULTIPROC_METHOD: spawn | ||
minReplicas: 1 | ||
maxReplicas: 1 | ||
targetRequests: 32 | ||
resourceProfile: nvidia-gpu-l4:1 |