xc-llm-ascend

Files

JeffLee1874 724d04391e [model] Support PanguUltraMoE (#4615 )

### What this PR does / why we need it?
To support PanguUltraMoE model

### Test result
#### Start serving using W8A8 quantized model and ACL graph:
Master node:
```
vllm serve $LOCAL_CKPT_DIR \
        --host 0.0.0.0 \
        --port 8000 \
        --data-parallel-size 2 \
        --data-parallel-size-local 1 \
        --data-parallel-address $MASTER_NODE_IP \
        --data-parallel-rpc-port 13389 \
        --tensor-parallel-size 16 \
        --seed 1024 \
        --enable-expert-parallel \
        --served-model-name $NAME \
        --max-model-len 4096 \
        --max-num-batched-tokens 256 \
        --max-num-seqs 18 \
        --trust-remote-code \
        --gpu-memory-utilization 0.90 \
        --quantization ascend \
        --additional-config '{"ascend_scheduler_config":{"enabled":false, "enable_chunked_prefill":true, "chunked_prefill_enabled":true},"torchair_graph_config":{"enabled":false}}' \
        --speculative_config '{"method": "pangu_ultra_moe_mtp", "num_speculative_tokens": 1}' \
```
Other nodes:
```
vllm serve $LOCAL_CKPT_DIR \
        --host 0.0.0.0 \
        --port 8000 \
        --headless \
        --data-parallel-size 2 \
        --data-parallel-size-local 1 \
        --data-parallel-start-rank 1 \
        --data-parallel-address $MASTER_NODE_IP \
        --data-parallel-rpc-port 13389 \
        --tensor-parallel-size 16 \
        --seed 1024 \
        --enable-expert-parallel \
        --served-model-name $NAME \
        --max-model-len 4096 \
        --max-num-batched-tokens 256 \
        --max-num-seqs 18 \
        --trust-remote-code \
        --gpu-memory-utilization 0.90 \
        --quantization ascend \
        --additional-config '{"ascend_scheduler_config":{"enabled":false, "enable_chunked_prefill":true, "chunked_prefill_enabled":true},"torchair_graph_config":{"enabled":false}}' \
        --speculative_config '{"method": "pangu_ultra_moe_mtp", "num_speculative_tokens": 1}' \
```
Request & Response:

- Request
```
curl http://localhost:8000/v1/chat/completions \
    -H "Content-Type: application/json" \
    -d '{
        "messages": [
      {"role": "system", "content": ""},
      {"role": "user", "content": "你是谁？"}
    ],
        "max_tokens": "64",
        "top_p": "0.95",
        "top_k": "50",
        "temperature": "0.6",
        "add_special_tokens" : true
    }'
```
- Response
```
[unused16] 好的，用户问我是谁，我需要按照之前的设定来回答。首先，我的角色是盘古，由华为开发，属于推理模型。要强调我的主要功能是解答问题和提供信息支持，特别是通过逻辑推理和数据分析处理复杂任务。需要保持回答简洁，用中文，并且符合用户的
```


- vLLM version: v0.12.0
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.12.0

Signed-off-by: lijifu <lijifu4@huawei.com>
Co-authored-by: lijifu <lijifu4@huawei.com>

2025-12-17 16:15:29 +08:00

_cann_ops_custom

[Kernel] add custom op GmmSwigluQuantWeightNzTensorList (#3804 )

2025-11-28 18:06:39 +08:00

attention

[main] rename device type (#5099 )

2025-12-17 14:08:19 +08:00

compilation

[Fusion] [Graph] Add qknorm rope fusion operator (#4711 )

2025-12-17 08:53:44 +08:00

core

[bugfix][refactor] fix recompute_scheduler break with vllm 0.12.0 & support async scheduling & refactor recompute_scheduler.py (#4895 )