ba3dfbd59e
### What this PR does / why we need it?
A refactoring of forward_context and model_runner_v1, add some context
which is necessary in model inference into forward_context, and refactor
dummy_run logic, make it more reasonable.
Some details for this PR:
Add `ascend_forward_context`;
Update mc2_v2 op, and support `active_mask` param;
Update scripts in examples dir;
refactor `dummy_run` logic;
Add soc_version for A2 and A3;
### Does this PR introduce _any_ user-facing change?
No change at user-facing.
### How was this patch tested?
- vLLM version: v0.10.0
- vLLM main:
57c22e57f9
Signed-off-by: zzzzwwjj <1183291235@qq.com>
34 lines
1.0 KiB
Bash
34 lines
1.0 KiB
Bash
rm -rf ./.torchair_cache/
|
|
rm -rf ./dynamo_*
|
|
rm -rf /root/ascend/log/debug/plog/*
|
|
|
|
export HCCL_IF_IP=2.0.0.0
|
|
export GLOO_SOCKET_IFNAME="enp189s0f0"
|
|
export TP_SOCKET_IFNAME="enp189s0f0"
|
|
export HCCL_SOCKET_IFNAME="enp189s0f0"
|
|
|
|
export OMP_PROC_BIND=false
|
|
export OMP_NUM_THREADS=100
|
|
|
|
export VLLM_USE_V1=1
|
|
export ASCEND_LAUNCH_BLOCKING=0
|
|
|
|
vllm serve /data/weights/Qwen2.5-0.5B-Instruct \
|
|
--host 0.0.0.0 \
|
|
--port 20002 \
|
|
--served-model-name Qwen \
|
|
--data-parallel-size 4 \
|
|
--data-parallel-size-local 4 \
|
|
--data-parallel-address 2.0.0.0 \
|
|
--data-parallel-rpc-port 13389 \
|
|
--tensor-parallel-size 4 \
|
|
--enable-expert-parallel \
|
|
--no-enable-prefix-caching \
|
|
--max-num-seqs 16 \
|
|
--max-model-len 4096 \
|
|
--max-num-batched-tokens 4096 \
|
|
--gpu-memory-utilization 0.9 \
|
|
--trust-remote-code \
|
|
--enforce-eager \
|
|
--additional-config '{"ascend_scheduler_config":{"enabled":true},"torchair_graph_config":{"enabled":false, "enable_multistream_moe":false, "use_cached_graph":false}}'
|