You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Make sure the doc url like `xxxx.html` works as before
- vLLM version: v0.23.0
- vLLM main:
vllm-project/vllm@1f486d9
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
Copy file name to clipboardExpand all lines: docs/source/tutorials/models/DeepSeek-R1.md
+3-3Lines changed: 3 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -169,7 +169,7 @@ Key Parameter Descriptions:
169
169
-`--no-enable-prefix-caching` indicates that prefix caching is disabled. To enable it, remove this option.
170
170
- If you use the w4a8 weight, more memory will be allocated to kvcache, and you can try to increase system throughput to achieve greater throughput.
171
171
172
-
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
172
+
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
173
173
174
174
Service Verification:
175
175
@@ -326,7 +326,7 @@ Key Parameter Descriptions:
326
326
-`--headless`: indicates that this vLLM instance is not the master service node. Only set on non-master nodes (Node 1). The master node (Node 0) should NOT set this flag.
327
327
- For single-node deployment, we recommend using `dp4 tp4` instead of `dp2 tp8`.
328
328
329
-
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
329
+
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
330
330
331
331
Service Verification:
332
332
@@ -441,4 +441,4 @@ This solution has been tested and demonstrates excellent performance.
441
441
442
442
## 10 FAQ
443
443
444
-
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/).
444
+
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html).
Copy file name to clipboardExpand all lines: docs/source/tutorials/models/DeepSeek-V3.1.md
+4-4Lines changed: 4 additions & 4 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -181,7 +181,7 @@ Key Parameter Descriptions:
181
181
-`--no-enable-prefix-caching` indicates that prefix caching is disabled. To enable it, remove this option.
182
182
- If you use the w4a8 weight, more memory will be allocated to kvcache, and you can try to increase system throughput to achieve greater throughput.
183
183
184
-
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
184
+
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
185
185
186
186
Service Verification:
187
187
@@ -355,7 +355,7 @@ Key Parameter Descriptions:
355
355
-`--headless`: indicates that this vLLM instance is not the master service node. Only set on non-master nodes (Node 1). The master node (Node 0) should NOT set this flag.
356
356
- For single-node deployment, we recommend using `dp4 tp4` instead of `dp2 tp8`.
357
357
358
-
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
358
+
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
359
359
360
360
Service Verification:
361
361
@@ -844,7 +844,7 @@ The proxy returns HTTP 200 OK. The JSON response contains the `choices` field wi
844
844
}
845
845
```
846
846
847
-
Common Issues Tip: If you encounter issues with PD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
847
+
Common Issues Tip: If you encounter issues with PD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
848
848
849
849
## 6 Functional Verification
850
850
@@ -965,4 +965,4 @@ Please refer to the [Feature Guide](../../user_guide/support_matrix/feature_matr
965
965
966
966
## 10 FAQ
967
967
968
-
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/).
968
+
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html).
Copy file name to clipboardExpand all lines: docs/source/tutorials/models/DeepSeek-V4-Flash.md
+3-3Lines changed: 3 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -235,7 +235,7 @@ Key Parameter Descriptions:
235
235
-`--async-scheduling` enables asynchronous scheduling to overlap CPU scheduling with NPU computation.
236
236
-`VLLM_ASCEND_ENABLE_FLASHCOMM1=1` enables the FlashComm communication optimization.
237
237
238
-
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
238
+
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
239
239
240
240
Service Verification:
241
241
@@ -839,7 +839,7 @@ Deployment Verification:
839
839
840
840
After the PD separation service is fully started, send a request through the proxy port on the prefill master node to verify that Prefill and Decode nodes are working correctly together. Refer to [Prefill-Decode Disaggregation (Deepseek)](../features/pd_disaggregation_mooncake_multi_node.md) for the proxy verification method.
841
841
842
-
Common Issues Tip: If you encounter issues withPD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
842
+
Common Issues Tip: If you encounter issues withPD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
843
843
844
844
#### 5.2.3 Ultra-Long Sequence Deployment
845
845
@@ -940,4 +940,4 @@ Please refer to the [Feature Guide](../../user_guide/support_matrix/feature_matr
940
940
941
941
## 10 FAQ
942
942
943
-
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/); this chapter only covers model-specific issues.
943
+
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html); this chapter only covers model-specific issues.
Copy file name to clipboardExpand all lines: docs/source/tutorials/models/DeepSeek-V4-Pro.md
+3-3Lines changed: 3 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -422,7 +422,7 @@ Key Parameter Descriptions:
422
422
-`--async-scheduling` enables asynchronous scheduling to overlap CPU scheduling with NPU computation.
423
423
-`VLLM_ASCEND_ENABLE_FLASHCOMM1=1` enables the FlashComm communication optimization.
424
424
425
-
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
425
+
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
426
426
427
427
Service Verification:
428
428
@@ -1169,7 +1169,7 @@ Deployment Verification:
1169
1169
1170
1170
After the PD separation service is fully started, send a request through the proxy port on the prefill master node to verify that Prefill and Decode nodes are working correctly together. Refer to [Prefill-Decode Disaggregation (Deepseek)](../features/pd_disaggregation_mooncake_multi_node.md) for the proxy verification method.
1171
1171
1172
-
Common Issues Tip: If you encounter issues withPD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
1172
+
Common Issues Tip: If you encounter issues withPD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
1173
1173
1174
1174
## 6 Functional Verification
1175
1175
@@ -1265,4 +1265,4 @@ Please refer to the [Feature Guide](../../user_guide/support_matrix/feature_matr
1265
1265
1266
1266
## 10 FAQ
1267
1267
1268
-
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/); this chapter only covers model-specific issues.
1268
+
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html); this chapter only covers model-specific issues.
Copy file name to clipboardExpand all lines: docs/source/tutorials/models/DeepSeekOCR2.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -236,7 +236,7 @@ Please refer to the [Feature Guide](../../user_guide/support_matrix/feature_matr
236
236
237
237
## 10 FAQ
238
238
239
-
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/).
239
+
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html).
240
240
241
241
-**Q: Startup fails with HCCL port conflicts (address already bound). What should I do?**
Copy file name to clipboardExpand all lines: docs/source/tutorials/models/Kimi-K2-Thinking.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -369,7 +369,7 @@ Please refer to the [Feature Guide](../../user_guide/support_matrix/feature_matr
369
369
370
370
## 10 FAQ
371
371
372
-
> For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/); this chapter only covers model-specific issues.
372
+
> For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html); this chapter only covers model-specific issues.
373
373
374
374
-**Q: API returns `{"error":"Model not found"}` or `404` when requesting with `model: "Kimi-K2-Thinking"`?**
Copy file name to clipboardExpand all lines: docs/source/tutorials/models/Kimi-K2.5.md
+4-4Lines changed: 4 additions & 4 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -179,7 +179,7 @@ Key Parameter Descriptions:
179
179
-`--mm-encoder-tp-mode` indicates how to optimize multi-modal encoder inference using tensor parallelism (TP). If you want to test the multimodal inputs, we recommend using `data`.
180
180
- If you use the w4a8 weight, more memory will be allocated to kvcache, and you can try to increase system throughput to achieve greater throughput.
181
181
182
-
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
182
+
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
183
183
184
184
Service Verification:
185
185
@@ -383,7 +383,7 @@ Key Parameter Descriptions:
383
383
-`--headless`: indicates that this vLLM instance is not the master service node. Only set on non-master nodes (Node 1). The master node (Node 0) should NOT set this flag.
384
384
- For single-node deployment, we recommend using `dp4 tp4` instead of `dp2 tp8`.
385
385
386
-
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
386
+
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
387
387
388
388
Service Verification:
389
389
@@ -903,7 +903,7 @@ The proxy returns HTTP 200 OK. The JSON response contains the `choices` field wi
903
903
}
904
904
```
905
905
906
-
Common Issues Tip: If you encounter issues with PD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
906
+
Common Issues Tip: If you encounter issues with PD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
907
907
908
908
## 6 Functional Verification
909
909
@@ -1040,7 +1040,7 @@ Please refer to the [Feature Guide](../../user_guide/support_matrix/feature_matr
1040
1040
1041
1041
## 10 FAQ
1042
1042
1043
-
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/); this chapter only covers model-specific issues.
1043
+
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html); this chapter only covers model-specific issues.
1044
1044
1045
1045
- **Q: What is the recommended TP/DP configuration for single-node deployment?**
Copy file name to clipboardExpand all lines: docs/source/tutorials/models/Kimi-K2.6.md
+3-3Lines changed: 3 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -183,7 +183,7 @@ Key Parameter Descriptions:
183
183
-`--mm-encoder-tp-mode` indicates how to optimize multi-modal encoder inference using tensor parallelism (TP). If you want to test the multimodal inputs, we recommend using `data`.
184
184
- If you use the w4a8 weight, more memory will be allocated to kvcache, and you can try to increase system throughput to achieve greater throughput.
185
185
186
-
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
186
+
Common Issues Tip: If you encounter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
187
187
188
188
Service Verification:
189
189
@@ -724,7 +724,7 @@ The proxy returns HTTP 200 OK. The JSON response contains the `choices` field wi
724
724
}
725
725
```
726
726
727
-
Common Issues Tip: If you encounter issues with PD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/) for troubleshooting.
727
+
Common Issues Tip: If you encounter issues with PD separation deployment, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html) for troubleshooting.
728
728
729
729
## 6 Functional Verification
730
730
@@ -867,7 +867,7 @@ Please refer to the [Feature Guide](../../user_guide/support_matrix/feature_matr
867
867
868
868
## 10 FAQ
869
869
870
-
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs/); this chapter only covers model-specific issues.
870
+
For common environment, installation, and general parameter issues, please refer to the [Public FAQ](https://docs.vllm.ai/projects/ascend/en/latest/faqs.html); this chapter only covers model-specific issues.
871
871
872
872
- **Q: What transformer version is required for tool_calls feature?**
0 commit comments