[Doc] Update accuracy reports for v0.10.0rc1

docs/source/developer_guide/evaluation/accuracy_report/Qwen2.5-VL-7B-Instruct.md

@@ -0,0 +1,21 @@
+# Qwen/Qwen2.5-VL-7B-Instruct
+
+**vLLM Version**: vLLM: 0.10.0 ([6d8d0a2](https://github.com/vllm-project/vllm/commit/6d8d0a2)),
+**vLLM Ascend Version**: v0.10.0rc1 ([4604882](https://github.com/vllm-project/vllm-ascend/commit/4604882))  
+**Software Environment**: CANN: 8.2.RC1, PyTorch: 2.7.1, torch-npu: 2.7.1.dev20250724  
+**Hardware Environment**: Atlas A2 Series  
+**Datasets**: mmmu_val  
+**Parallel Mode**: TP  
+**Execution Mode**: ACLGraph  
+
+**Command**:  
+
+```bash
+export MODEL_ARGS='pretrained=Qwen/Qwen2.5-VL-7B-Instruct,tensor_parallel_size=1,dtype=auto,trust_remote_code=False,max_model_len=8192'
+lm_eval --model vllm-vlm --model_args $MODEL_ARGS --tasks mmmu_val \
+--apply_chat_template True --fewshot_as_multiturn True  \
+--limit None --batch_size auto
+```
+| Task                  | Metric      | Value     | Stderr |
+|-----------------------|-------------|----------:|-------:|
+|                mmmu_val |        acc,none |✅0.5211111111111111 | ± 0.0162 |

docs/source/developer_guide/evaluation/accuracy_report/Qwen3-30B-A3B.md

@@ -0,0 +1,22 @@
+# Qwen/Qwen3-30B-A3B
+
+**vLLM Version**: vLLM: 0.10.0 ([6d8d0a2](https://github.com/vllm-project/vllm/commit/6d8d0a2)),
+**vLLM Ascend Version**: v0.10.0rc1 ([4604882](https://github.com/vllm-project/vllm-ascend/commit/4604882))  
+**Software Environment**: CANN: 8.2.RC1, PyTorch: 2.7.1, torch-npu: 2.7.1.dev20250724  
+**Hardware Environment**: Atlas A2 Series  
+**Datasets**: gsm8k  
+**Parallel Mode**: TP  
+**Execution Mode**: ACLGraph  
+
+**Command**:  
+
+```bash
+export MODEL_ARGS='pretrained=Qwen/Qwen3-30B-A3B,tensor_parallel_size=2,dtype=auto,trust_remote_code=False,max_model_len=4096,gpu_memory_utilization=0.6,enable_expert_parallel=True'
+lm_eval --model vllm --model_args $MODEL_ARGS --tasks gsm8k \
+--apply_chat_template False --fewshot_as_multiturn False  --num_fewshot 5  \
+--limit None --batch_size auto
+```
+| Task                  | Metric      | Value     | Stderr |
+|-----------------------|-------------|----------:|-------:|
+|                   gsm8k | exact_match,strict-match |✅0.8938589840788476 | ± 0.0085 |
+|                   gsm8k | exact_match,flexible-extract |✅0.8476118271417741 | ± 0.0099 |

docs/source/developer_guide/evaluation/accuracy_report/Qwen3-8B-Base.md

@@ -0,0 +1,22 @@
+# Qwen/Qwen3-8B-Base
+
+**vLLM Version**: vLLM: 0.10.0 ([6d8d0a2](https://github.com/vllm-project/vllm/commit/6d8d0a2)),
+**vLLM Ascend Version**: v0.10.0rc1 ([4604882](https://github.com/vllm-project/vllm-ascend/commit/4604882))  
+**Software Environment**: CANN: 8.2.RC1, PyTorch: 2.7.1, torch-npu: 2.7.1.dev20250724  
+**Hardware Environment**: Atlas A2 Series  
+**Datasets**: gsm8k  
+**Parallel Mode**: TP  
+**Execution Mode**: ACLGraph  
+
+**Command**:  
+
+```bash
+export MODEL_ARGS='pretrained=Qwen/Qwen3-8B-Base,tensor_parallel_size=1,dtype=auto,trust_remote_code=False,max_model_len=4096'
+lm_eval --model vllm --model_args $MODEL_ARGS --tasks gsm8k \
+--apply_chat_template True --fewshot_as_multiturn True  --num_fewshot 5  \
+--limit None --batch_size auto
+```
+| Task                  | Metric      | Value     | Stderr |
+|-----------------------|-------------|----------:|-------:|
+|                   gsm8k | exact_match,strict-match |✅0.8278999241849886 | ± 0.0104 |
+|                   gsm8k | exact_match,flexible-extract |✅0.8294162244124337 | ± 0.0104 |

docs/source/developer_guide/evaluation/accuracy_report/index.md

@@ -0,0 +1,9 @@
+# Accuracy Report
+
+:::{toctree}
+:caption: Accuracy Report
+:maxdepth: 1
+Qwen2.5-VL-7B-Instruct
+Qwen3-30B-A3B
+Qwen3-8B-Base
+:::