【docs】【NPU】Update Expert Parallelism docs for Ascend NPU (#17940)
This commit is contained in:
@@ -27,7 +27,7 @@ The support matrix is split into two parts: MHA (standard attention) and MLA (mu
|
||||
| **Dual Chunk FlashAttention** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
||||
| **AITER (ROCm)** | ✅ | ✅ | ❌ | ✅ | ✅ | ❌ | ✅ |
|
||||
| **Wave (ROCm)** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
||||
| **Ascend (NPU)** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ |
|
||||
| **Ascend (NPU)** | ✅ | ❌ | ❌ | ✅ | ❌ | ❌ | ✅ |
|
||||
| **Intel XPU** | ✅ | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ |
|
||||
| **Intel AMX (CPU)** | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
||||
|
||||
|
||||
Reference in New Issue
Block a user