[Test] Add nightly test for DeepSeek-V3.2-Exp #3908
+306
−45
There are no files selected for viewing
This file contains hidden or 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 hidden or 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 hidden or 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 +1,105 @@ | ||
| # Copyright (c) 2025 Huawei Technologies Co., Ltd. All Rights Reserved. | ||
| # Copyright 2023 The vLLM team. | ||
| # | ||
| # Licensed under the Apache License, Version 2.0 (the "License"); | ||
| # you may not use this file except in compliance with the License. | ||
| # You may obtain a copy of the License at | ||
| # | ||
| # http://www.apache.org/licenses/LICENSE-2.0 | ||
| # | ||
| # Unless required by applicable law or agreed to in writing, software | ||
| # distributed under the License is distributed on an "AS IS" BASIS, | ||
| # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | ||
| # See the License for the specific language governing permissions and | ||
| # limitations under the License. | ||
| # This file is a part of the vllm-ascend project. | ||
| # | ||
| from typing import Any | ||
|
|
||
| import openai | ||
| import pytest | ||
| from vllm.utils import get_open_port | ||
|
|
||
| from tests.e2e.conftest import RemoteOpenAIServer | ||
| from tools.aisbench import run_aisbench_cases | ||
|
|
||
| MODELS = [ | ||
| "vllm-ascend/DeepSeek-V3.2-Exp-W8A8", | ||
| ] | ||
|
|
||
| TENSOR_PARALLELS = [8] | ||
| DATA_PARALLELS = [2] | ||
|
|
||
| prompts = [ | ||
| "San Francisco is a", | ||
| ] | ||
|
|
||
| api_keyword_args = { | ||
| "max_tokens": 10, | ||
| } | ||
|
|
||
| aisbench_cases = [{ | ||
| "case_type": "accuracy", | ||
| "dataset_path": "vllm-ascend/gsm8k-lite", | ||
| "request_conf": "vllm_api_general_chat", | ||
| "dataset_conf": "gsm8k/gsm8k_gen_0_shot_cot_chat_prompt", | ||
| "max_out_len": 32768, | ||
| "batch_size": 32, | ||
| "baseline": 95, | ||
| "threshold": 5 | ||
| }, { | ||
| "case_type": "performance", | ||
| "dataset_path": "vllm-ascend/GSM8K-in3500-bs400", | ||
| "request_conf": "vllm_api_stream_chat", | ||
| "dataset_conf": "gsm8k/gsm8k_gen_0_shot_cot_str_perf", | ||
| "num_prompts": 80, | ||
| "max_out_len": 1500, | ||
| "batch_size": 20, | ||
| "request_rate": 0, | ||
| "baseline": 1, | ||
| "threshold": 0.97 | ||
| }] | ||
|
|
||
|
|
||
| @pytest.mark.asyncio | ||
| @pytest.mark.parametrize("model", MODELS) | ||
| @pytest.mark.parametrize("tp_size", TENSOR_PARALLELS) | ||
| @pytest.mark.parametrize("dp_size", DATA_PARALLELS) | ||
| async def test_models(model: str, tp_size: int, dp_size: int) -> None: | ||
| port = get_open_port() | ||
| env_dict = { | ||
| "TASK_QUEUE_ENABLE": "1", | ||
| "OMP_PROC_BIND": "false", | ||
| "HCCL_OP_EXPANSION_MODE": "AIV", | ||
| "PAGED_ATTENTION_MASK_LEN": "5500", | ||
| "DYNAMIC_EPLB": "true" | ||
| } | ||
| server_args = [ | ||
| "--no-enable-prefix-caching", "--enable-expert-parallel", | ||
| "--tensor-parallel-size", | ||
| str(tp_size), "--data-parallel-size", | ||
| str(dp_size), "--port", | ||
| str(port), "--max-model-len", "36864", "--max-num-batched-tokens", | ||
| "36864", "--block-size", "128", "--trust-remote-code", | ||
| "--quantization", "ascend", "--gpu-memory-utilization", "0.9", | ||
| "--additional-config", '{"ascend_scheduler_config":{"enabled":true},' | ||
| '"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}' | ||
| ] | ||
| request_keyword_args: dict[str, Any] = { | ||
| **api_keyword_args, | ||
| } | ||
| with RemoteOpenAIServer(model, | ||
| server_args, | ||
| server_port=port, | ||
| env_dict=env_dict, | ||
| auto_port=False) as server: | ||
| client = server.get_async_client() | ||
| batch = await client.completions.create( | ||
| model=model, | ||
| prompt=prompts, | ||
| **request_keyword_args, | ||
| ) | ||
| choices: list[openai.types.CompletionChoice] = batch.choices | ||
| assert choices[0].text, "empty response" | ||
| # aisbench test | ||
| run_aisbench_cases(model, port, aisbench_cases) | ||
This file contains hidden or 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 +1,60 @@ | ||||
| test_name: "test DeepSeek-V3.2-Exp-bf1 disaggregated_prefill" | ||||
| model: "Yanguan/DeepSeek-V3.2-Exp-bf16" | ||||
| num_nodes: 2 | ||||
| npu_per_node: 16 | ||||
| env_common: | ||||
| VLLM_USE_MODELSCOPE: true | ||||
| OMP_PROC_BIND: false | ||||
| OMP_NUM_THREADS: 100 | ||||
| HCCL_BUFFSIZE: 1024 | ||||
| SERVER_PORT: 8080 | ||||
| disaggregated_prefill: | ||||
|
There was a problem hiding this comment. I think you want to add multi-dp test cross 2 nodes for deepseekv3.2 right? so we should not pass disaggregated_prefill related config |
||||
| enabled: true | ||||
| prefiller_host_index: [0] | ||||
| decoder_host_index: [1] | ||||
|
|
||||
| deployment: | ||||
| - | ||||
| server_cmd: > | ||||
| vllm serve Yanguan/DeepSeek-V3.2-Exp-bf16 \ | ||||
| --host 0.0.0.0 | ||||
| --port $SERVER_PORT | ||||
| --data-parallel-address $LOCAL_IP | ||||
| --data-parallel-size 2 | ||||
| --data-parallel-size-local 1 | ||||
| --data-parallel-rpc-port 13389 | ||||
| --tensor-parallel-size 16 | ||||
| --seed 1024 | ||||
| --served-model-name deepseek_v3.2 | ||||
|
There was a problem hiding this comment.
Suggested change
There was a problem hiding this comment. Since we use the model name instead of this short tag when checking if the service is running, passing this parameter will result in a timeout. |
||||
| --enable-expert-parallel | ||||
| --max-num-seqs 16 | ||||
| --max-model-len 17450 | ||||
| --max-num-batched-tokens 17450 | ||||
| --trust-remote-code | ||||
| --no-enable-prefix-caching | ||||
| --gpu-memory-utilization 0.9 | ||||
| --additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}' | ||||
|
|
||||
| - | ||||
| server_cmd: > | ||||
| vllm serve Yanguan/DeepSeek-V3.2-Exp-bf16 \ | ||||
| --host 0.0.0.0 | ||||
| --port $SERVER_PORT | ||||
| --headless | ||||
| --data-parallel-size 2 | ||||
| --data-parallel-size-local 1 | ||||
| --data-parallel-start-rank 1 | ||||
| --data-parallel-address $MASTER_IP | ||||
| --data-parallel-rpc-port 13389 | ||||
| --tensor-parallel-size 16 | ||||
| --seed 1024 | ||||
| --served-model-name deepseek_v3.2 | ||||
|
|
||||
| --max-num-seqs 16 | ||||
| --max-model-len 17450 | ||||
| --max-num-batched-tokens 17450 | ||||
| --enable-expert-parallel | ||||
| --trust-remote-code | ||||
| --no-enable-prefix-caching | ||||
| --gpu-memory-utilization 0.92 | ||||
| --additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":true,"graph_batch_sizes":[16]}}' | ||||
zhangxinyuehfad marked this conversation as resolved.
Show resolved
Hide resolved
|
||||
| benchmarks: | ||||
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Uh oh!
There was an error while loading. Please reload this page.