Commit Graph
1991 Commits
Author SHA1 Message Date
Nengjun Ma 48811bc0b8 Optimize the print info format when deprecated code is used in vllm-ascend (#5696)
### What this PR does / why we need it?
Optimize the warning print information format when detects depredated
code is used in vllm-ascend.

### Does this PR introduce _any_ user-facing change?
NA

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d

---------

Signed-off-by: leo-pony <nengjunma@outlook.com>
2026-01-08 09:26:49 +08:00
Aoxuan Chen 8763953f56 [Feature] add the magicmtp speculative decoding acceleration algorithm (#5542)
### What this PR does / why we need it?

1. MagicMTP (paper: "Block Verification Accelerates Speculative
Decoding") was introduced to consider the influence among multiple draft
tokens, improving the acceptance rate without compromising accuracy.
2. Added Triton and PyTorch implementations, and added E2E test cases.

### Does this PR introduce _any_ user-facing change?
MagicMTP will automatically take effect when the parameter
"num_speculative_tokens" >= 3.
- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: chenaoxuan <cax1165@163.com>
2026-01-08 09:15:55 +08:00
lidenghui1110 481138e1d2 [bugfix] adapt to new implemented get_kv_cache_spec in cpuoffload connector (#4311)
### What this PR does / why we need it?
func `get_kv_cache_spec` in model_runner changed a lot and caused error
in cpuoffloading connector which is copied from model_runner, this PR
adapts to new implemented `get_kv_cache_spec` to fix it.

### How was this patch tested?

- vLLM version: v0.11.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2918c1b49c88c29783c86f78d2c4221cb9622379

Signed-off-by: lidenghui <lidenghui1110@gmail.com>
2026-01-08 09:15:09 +08:00
f7db812ed7 [refactor] Refactor the interface for shard weight and remove the flashcomm2 o_shared interface. (#5181)
### What this PR does / why we need it?
- Delete the environment variable
`VLLM_ASCEND_ENABLE_FLASHCOMM2_OSHARED`
- Introduce layer_sharding as a configurable feature in
additional_config
- Revise the term "shared weight" to "shard weight."
Configuration : The feature is opt-in via the additional_config
argument:
```
--additional-config '{
  "layer_sharding": ["o_proj", "q_b_proj"]
}'
```

This is orthogonal to standard tensor parallelism and weight replication
strategies. It is treated as a separate, explicit feature.It can be used
in any scenario, combined with the
flashcomm2https://github.com/vllm-project/vllm-ascend/pull/3232 feature
or the ShardedCP #4702 feature, to achieve significant performance.



- vLLM version: v0.12.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/ad32e3e19ccf0526cb6744a5fed09a138a5fb2f9

---------

Signed-off-by: zzhx1 <zzh_201018@outlook.com>
Signed-off-by: zzhxx <zhangzihang23@mails.ucas.ac.cn>
Signed-off-by: chenxiao <Jaychou1620@Gmail.com>
Co-authored-by: clrs97 <524936896@qq.com>
Co-authored-by: Levi-JQ <yujinqi2@huawei.com>
Co-authored-by: chenxiao <Jaychou1620@Gmail.com>
2026-01-08 09:05:02 +08:00
zxr2333andliziyu 20a8cf061b [BugFix][P/D] Fix pre-create link parameter error (#5694)
### What this PR does / why we need it?
Fix pre-create link parameter error, `batch_transfer_sync_write`
requires list.

### Does this PR introduce _any_ user-facing change?
No.

### How was this patch tested?
By CI.

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d

Signed-off-by: liziyu <liziyu16@huawei.com>
Co-authored-by: liziyu <liziyu16@huawei.com>
2026-01-08 08:41:10 +08:00
ZCG12345 3be8e33fe9 [Kernel] Add moe_gating_top_k operator support for Ascend NPU (#5579)
### What this PR does / why we need it?

1.replace moe_gating_top_k from torch_npu with custom op
2.enable the  renorm function of moe_gating_top_k in softmax scenerio

### Does this PR introduce _any_ user-facing change?
No

### How was this patch tested?
No need test

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

---------

Signed-off-by: ZCG12345 <2097562023@qq.com>
2026-01-07 21:42:31 +08:00
Li Wang 1165b2c863 [1/N][CI] Refactor accuracy test (#5400)
### What this PR does / why we need it?
1. Accuracy testing no longer compares eager and graph modes; instead,
it directly extracts the golden result under the graph mode
configuration (the implicit purpose of this case is to verify whether
modifications affect existing results)
2. Next step: finer-grained supervision of logits/sampler results
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: release/v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/254f6b986720c92ddf97fbb1a6a6465da8e87e29

Signed-off-by: wangli <wangli858794774@gmail.com>
2026-01-07 20:58:15 +08:00
Icey b94fc13d3f [BugFix][Fusion] Fix graph fusion failure problem (#5676)
Currently, the vllm pull request
(https://github.com/vllm-project/vllm/pull/24252) is causing operator
fusion to fail. This issue was previously fixed by patching the backend.
The root cause has been identified, and the problem can be resolved with
this pull request.
- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d

---------

Signed-off-by: wxsIcey <1790571317@qq.com>
2026-01-07 18:42:55 +08:00
Icey 137f28341d [Tests] Add qwen3-8b nightly test (#5597)
### What this PR does / why we need it?
Add qwen3-8b nightly test 

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731
---------
Signed-off-by: wxsIcey <1790571317@qq.com>
2026-01-07 18:42:05 +08:00
Mengqing Cao 3f4f2b4ae6 [Refactor] Import global var form vllm instead of overwirte it (#5469)
### What this PR does / why we need it?
Import global var form vllm instead of overwirte it, so that we could
use the correct global variant value

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/5326c89803566a131c928f7fdd2100b75c981a42
---------
Signed-off-by: MengqingCao <cmq0113@163.com>
2026-01-07 18:41:45 +08:00
LICO67373andweijinqian0 380f089fbf [Refactor] Fix AttentionMaskBuilder singleton and remove redundant pcp_prefill_mask (#4870)
## What this PR does / why we need it?

This PR fixes the `AttentionMaskBuilder` singleton initialization issue
introduced in PR #4779 and removes the unused `pcp_prefill_mask` field.

### Background

After PR #4779 made `AttentionMaskBuilder` a singleton with `@singleton`
decorator, the class constructor now requires a `device` parameter.
However, two initialization sites were still using the old parameterless
constructor, causing failures.

### Changes

1. **Fix singleton initialization**
- Fixed `AttentionMaskBuilder()` → `AttentionMaskBuilder(self.device)`
in `AscendMLAMetadataBuilder.__init__()`
- Fixed `AttentionMaskBuilder()` → `AttentionMaskBuilder(self.device)`
in `AscendAttentionMetadataBuilder.__init__()`

2. **Remove unused field**
- Removed `pcp_prefill_mask` field from
`AscendPrefillContextParallelMetadata` (never used in codebase)
   - Updated related test assertions

### Related

- Issue #5463
- PR #4779 (Unify all mask generation methods)
- PR #5389 (Make AttentionMaskBuilder singleton)

## Does this PR introduce _any_ user-facing change?

No. This is an internal refactoring.

## How was this patch tested?

- ✅ Local testing: No linter errors
- ✅ Unit tests for attention modules verified
- ⏳ CI pipeline

Signed-off-by: lico67373 <918688502@qq.com>
Co-authored-by: weijinqian0 <1184188277@qq.com>
2026-01-07 17:09:52 +08:00
wangxiyuan 91790fd85a [CI] move image and wheel job to schedule way (#5685)
move image and wheel job to schedule way to save CI resource

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
2026-01-07 16:40:19 +08:00
无脸男 1140789e83 [Bugfix] Fix the graph capture failure issue in the eagle3+full scenario. (#5553)
### What this PR does / why we need it?
When launching the service in the scenario where the
cudagraph_mode is set to FULL and Eagle3 acceleration is enabled for
inference, an error in fia will cause graph capture to fail. This PR
fixes the issue.

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: WithHades <244036962@qq.com>
2026-01-07 15:57:16 +08:00
weiguihua2 2b8a9ce8bd [Bugfix] fix resource are insufficient when pcp and piecewise (#5377)
### What this PR does / why we need it?
Resolving the issue of insufficient resources during service operation
when PCP is enabled in a piecewise scenario.

When enabling PCP and executing in piecewise mode, the curl request
fails due to insufficient resources, resulting in the error message "The
resources are insufficient." Through profiling analysis, it was found
that the PCP communication domain also occupies streams and consumes
resources. Therefore, when updating aclgraph sizes, the PCP
communication domain needs to be taken into account.

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: release/v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/bc0a5a0c089844b17cb93f3294348f411e523586

---------

Signed-off-by: weiguihua2 <weiguihua2@huawei.com>
2026-01-07 15:39:52 +08:00
Paco Xu 4f9808002b [CI] Add workflow to cancel running workflows on PR close (#5646)
Example:
https://github.com/vllm-project/vllm-ascend/actions/runs/20735955959/job/59533181655
is still running after
https://github.com/vllm-project/vllm-ascend/pull/5612 is closed.

And the action will be running for more than 2 hours, which needs to be
cleanup.

It seems that the Github Aciton will not cancel it automatically, so I
add this to cannel those PR related actions once it is closed.

Tested in https://github.com/pacoxu/pacoxu/actions/runs/20743173119.
- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d

Signed-off-by: Paco Xu <roollingstone@gmail.com>
2026-01-07 15:38:10 +08:00
Li Wang d314ea8d3d [CI] Bump lm-eval version to v0.4.9.2 (#5655)
### What this PR does / why we need it?
fix https://github.com/vllm-project/vllm-ascend/issues/2865, lm-eval [got an official update last
month](https://github.com/EleutherAI/lm-evaluation-harness/releases/tag/v0.4.9.2),
so let's bump the version.

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d

Signed-off-by: wangli <wangli858794774@gmail.com>
2026-01-07 14:15:53 +08:00
wangxiyuan 6f7a81cd9f [CI] cleanup single/multi-card test (#5623)
1. speed up e2e light test.
2. create `2-cards` and `4-cards` folder in multicard
3. move ops to nightly
4. run test in Alphabetical Order

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/8be6432bdaf6275664d857b1e5e9bf8ed1ce299e

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
2026-01-07 14:13:34 +08:00
SILONG ZENGandwangli 1afbc01ed4 [misc]Add Kimi-K2 series to CI model list (#5656)
### What this PR does / why we need it?
Add the model to CI for subsequent addition of nightly test cases:
- moonshotai/Kimi-K2-Thinking
- vllm-ascend/Kimi-K2-Instruct-W8A8

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d

---------

Signed-off-by: MrZ20 <2609716663@qq.com>
Signed-off-by: wangli <wangli858794774@gmail.com>
Co-authored-by: wangli <wangli858794774@gmail.com>
2026-01-07 11:32:48 +08:00
UnifiedCacheManager d6bb17f10e [Bugfix]Add register_kv_cache in ucm_connector (#5657)
### What this PR does / why we need it?
To adapt different shapes of the KV cache, UCM optimized the
initialization of store by moving it into `register_kv_caches`.
Therefore, this update adds `register_kv_caches` interface to
UCMConnectorV1.

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d

Signed-off-by: UnifiedCacheManager <unifiedcachem@163.com>
2026-01-07 11:30:33 +08:00
LI SHENGYONG cd59323e40 [Bugfix] Revert pr4214 multi-stream collect expert hotpot (#5529)
### What this PR does / why we need it?
PR4214 was intended to collect expert heat by processing multiple
streams, which could lead to memory overwriting and accuracy issues.
After communicating with the PR submitter, this PR has been reverted.

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?
qwen3-moe dynamic eplb
Befor revert
| dataset | version | metric | mode | vllm-api-general-chat |
|----- | ----- | ----- | ----- | -----|
| aime2024 | 604a78 | accuracy | gen | 43.33 |

After revert 
| dataset | version | metric | mode | vllm-api-general-chat |
|----- | ----- | ----- | ----- | -----|
| aime2024 | 604a78 | accuracy | gen | 86.67 |

baseline (without eplb)
| dataset | version | metric | mode | vllm-api-general-chat |
|----- | ----- | ----- | ----- | -----|
| aime2024 | 604a78 | accuracy | gen | 86.67 |
- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/45c1ca1ca1ee8fa06df263c8715e8a412ff408d4

Signed-off-by: shenchuxiaofugui <1311027364@qq.com>
2026-01-07 11:26:47 +08:00
wangyibo1005 25baf6df09 [Feature]EPLB:Adapt DispatchGmmCombineDecode operator to eplb tensor list and expert token numbers (#5552)
#### What this PR does / why we need it?
This PR adapt DispatchGmmCombineDecode operator to eplb tensor list and
expert token numbers.

This operator support gmm1, gmm2, gmm1Scale and gmm2Scale in format of
list.
This operator support couting how many token each local expert recieves
by expertTokensNum .


- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

More info about this operator, please refer to RFC: issue
https://github.com/vllm-project/vllm-ascend/issues/5476
2026-01-07 11:23:42 +08:00
starmountain1997andguozr 086c093347 [CI] Add DeepSeek-V3.2-W8A8 nightly ci test (#5371)
# What this PR does / why we need it?

Add DeepSeek-V3.2-W8A8 dual-node nightly CI test and update A3 nightly
test configuration:

1. Add DeepSeek-V3.2-W8A8 dual-node test:
tests/e2e/nightly/multi_node/config/DeepSeek-V3_2-W8A8-A3-dual-nodes.yaml
    - 2 nodes, 16 NPUs per node (32 NPUs total)
- Configuration: 2P+1D (data-parallel-size=4, tensor-parallel-size=8,
data-parallel-size-local=2)
    - Includes performance and accuracy benchmarks with GSM8K dataset
  2. Update A3 nightly workflow: .github/workflows/nightly_test_a3.yaml
- Added DeepSeek-V3.2-W8A8 dual-node test to the A3 nightly test matrix
    - Test name: multi-node-dpsk3.2-2node
3. Improve test scripts: Updated
.github/workflows/_e2e_nightly_multi_node.yaml and related scripts for
better multi-node testing support

test on A3 instances
  - Performance baseline: 1 (threshold: 0.97)
  - Accuracy baseline: 95% (threshold: 5%)
- Test dataset: GSM8K with 512 prompts for performance, gsm8k-lite for
accuracy
---------
Signed-off-by: guozr <guozr1997@hotmail.com>
Co-authored-by: guozr <guozr1997@hotmail.com>
2026-01-07 10:02:02 +08:00
Feng LiuandF.Liu cbc987db0b [bugfix (pcp)] fix chunked prefill accurancy issue (#5647)
### What this PR does / why we need it?
Purpose: initialize padded slot mapping buffer to prevent garbage
values.

In PCP mode, the `pcp_padded_slot_mapping` buffer is reused across
invocations. Without explicit initialization, this buffer retain stale
values from previous runs, which can lead to incorrect results.

This change ensures the buffer is filled with -1.

### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d

---------

Signed-off-by: F.Liu <liufeng248@huawei.com>
Co-authored-by: F.Liu <liufeng248@huawei.com>
2026-01-07 10:01:27 +08:00
wangxiyuan 1112208052 [Refactor] Cleanup platform (#5566)
### What this PR does / why we need it?
1. add `COMPILATION_PASS_KEY` constant
2. clean up useless platform interface `empty_cache`, `synchronize`,
`mem_get_info`, `clear_npu_memory`
3. rename `CUSTOM_OP_REGISTERED` to `_CUSTOM_OP_REGISTERED`
4. remove uesless env `VLLM_ENABLE_CUDAGRAPH_GC`

NPUPlatform is the interface called by vLLM. Do not call it inner
vllm-ascend.

### Does this PR introduce _any_ user-facing change?
This PR is just  a cleanup. All CI should pass.

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
2026-01-07 09:25:55 +08:00
6ea2afe5fa [Feature] implement basic framework for batch invariant (#5517)
### What this PR does / why we need it?
This PR implement the basic framework for batch invariant, please see
https://github.com/vllm-project/vllm-ascend/issues/5487.
### Does this PR introduce _any_ user-facing change?
we reuse the function `vllm_is_batch_invariant` in vllm to judge if
batch invariant is enabled.

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/45c1ca1ca1ee8fa06df263c8715e8a412ff408d4
---------
Signed-off-by: Ronald1995 <ronaldautomobile@163.com>
Signed-off-by: Lord_of_Ironhill <suiweiyi@huawei.com>
Signed-off-by: zjchenn <zjchenn@gmail.com>
Signed-off-by: wangx700 <wangxin700@huawei.com>
Co-authored-by: Lord_of_Ironhill <suiweiyi@huawei.com>
Co-authored-by: zjchenn <zjchenn@gmail.com>
Co-authored-by: wangx700 <wangxin700@huawei.com>
2026-01-07 09:11:26 +08:00
CodeCat bdedf3c9f8 [Graph][Fusion] Add AddRMSNormSPPattern and AddRMSNormSPPatternWithBias (#5569)
### What this PR does / why we need it?
This PR builds upon PR
https://github.com/vllm-project/vllm-ascend/pull/5011 and aims to
further enhance the npu_graph_ex_passes module. Based on prior work, we
have added graph optimization support for the add_rms_quant fused
operator in scenarios where a bias term is present—ensuring the fusion
pattern is correctly registered and matched into the computation graph.

For validation, we switched to the Qwen3-235B-A22B-W8A8 model for
SPPatternWithBias and Qwen3-32B model for SPPattern. Benchmark results
show that, compared to the unfused baseline, enabling this fusion pass
significantly improves inference throughput for W8A8 quantized models.
For more details can refer to the
RFC:https://github.com/vllm-project/vllm-ascend/issues/4715
### Does this PR introduce _any_ user-facing change?
no
### How was this patch tested?
```
llm = LLM(
        model=model,
        tensor_parallel_size=GPUs_per_dp_rank,
        enforce_eager=False,
        enable_expert_parallel=enable_expert_parallel,
        trust_remote_code=trust_remote_code,
        gpu_memory_utilization=0.98,
        max_num_batched_tokens=512,
        # load_format="dummy",
        max_model_len=2048,
        max_num_seqs=16,
        quantization="ascend",
        additional_config={
            "refresh": True,
            "enable_npugraph_ex": True
        },
        compilation_config={
            "cudagraph_capture_sizes": [8, 16],
            "cudagraph_mode": "FULL_DECODE_ONLY",
        },
    )
    if profile_dir:
        llm.start_profile()
    outputs = llm.generate(prompts, sampling_params)
    if profile_dir:
        llm.stop_profile()
    for i, output in enumerate(outputs):
        if i >= 5:
            break
        prompt = output.prompt
        generated_text = output.outputs[0].text
        print(
            f"DP rank {global_dp_rank}, Prompt: {prompt!r}, "
            f"Generated text: {generated_text!r}"
        )
```
- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: cjian <2318164299@qq.com>
2026-01-07 09:03:45 +08:00
zhenwenqi2024 ad9b711f89 [Bugfix] fix dcp_only bug and add e2e accuracy test for dcp only and pcp only (#5565)
### What this PR does / why we need it?
[Bugfix] fix dcp_only bug and add e2e accuracy test for dcp only and pcp
only
this pr fix the bug of accuracy test when decode_parallel_size>1 and
prefill_context_parallel_size=1.
### Does this PR introduce _any_ user-facing change?
NO

### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

---------

Signed-off-by: zhenwenqi2024 <zhenwenqi_2022@qq.com>
2026-01-06 22:48:21 +08:00
Fager10086 77a029979e Revert "[BugFix][Fusion] Fix graph fusion failure problem (#5253)" (#5667)
### What this PR does / why we need it?

Revert PR 5253 to fix the smoking problem

### Does this PR introduce _any_ user-facing change?

Does not.

### How was this patch tested?

It was tested in the failure case.

Signed-off-by: Rifa <865071616@qq.com>
2026-01-06 21:55:47 +08:00
330e25ab1d [P/D] Performance enhancement of Layerwise connector in TP asymmetric scenarios (#5540)
### What this PR does / why we need it?
[P/D] Performance enhancement of Layerwise connector in TP asymmetric
scenarios
1. Session fusion: For transmission tasks at each layer, aggregate
transmission tasks with the same destination and merge them into a
single task for assignment.
2. Alltoall aggregation: For TP asymmetric scenarios, perform all
alltoall operations at once according to the block granularity for all
requests.

[RFC]: CDCP Scheduling for Disaggregated Prefilling with KV Cache
Layerwise Push Support
https://github.com/vllm-project/vllm-ascend/issues/4842
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/45c1ca1ca1ee8fa06df263c8715e8a412ff408d4

---------

Signed-off-by: liziyu <liziyu16@huawei.com>
Signed-off-by: nwpu-zxr <zhouxuerong2@huawei.com>
Signed-off-by: wangxiaoteng <wangxiaoteng@huawei.com>
Co-authored-by: nwpu-zxr <zhouxuerong2@huawei.com>
Co-authored-by: wangxiaoteng <wangxiaoteng@huawei.com>
2026-01-06 20:25:36 +08:00
wangxiyuan cd1162e25a [Misc] Remove useless weight loader patch (#5619)
The patch for weight loader is useless now. Let's remove it

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/8be6432bdaf6275664d857b1e5e9bf8ed1ce299e

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
2026-01-06 20:17:32 +08:00
InSec 089ca2ddcc [Nightly][Test] Add Qwen3-Next-80B-A3B-Instruct-W8A8 nightly test (#5616)
### What this PR does / why we need it?
There was an accuracy issue with **Qwen3-Next-80B-A3B-Instruct-W8A8**
model in the old version of **Triton-Ascend**, so, we are now adding one
nightly test to maintain it.
### Does this PR introduce _any_ user-facing change?
N/A
### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: IncSec <1790766300@qq.com>
2026-01-06 17:36:00 +08:00
yeyifanandMengqing Cao cc0110abb4 [Bugfix] Remove swa parameter of fia (#5602)
### What this PR does / why we need it?
When using the swa parameter in fia, headDim does not currently support
256, and when gemma3's headDim is equal to 256, an error will occur.
Therefore, code rollback is required, and it will be incorporated after
cann supports it.
### Does this PR introduce _any_ user-facing change?
Remove swa parameter of fia.
### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

---------

Signed-off-by: nsdie <yeyifan@huawei.com>
Co-authored-by: Mengqing Cao <cmq0113@163.com>
2026-01-06 17:24:43 +08:00
29e2f9a43e Bugfix: Align expert map shapes with redundant experts in EPLB adjustment (#5285)
#### Overview
This PR fixes a shape mismatch bug between `expert_placement_map` and
`log2phy_expert_map` when **redundant experts** are enabled in the
vLLM-Ascend platform. The issue occurred during the initialization of
expert maps and their updates via EPLB (Expert Load Balancer)
adjustment, leading to potential tensor shape errors and incorrect
expert routing in distributed MoE deployments.

#### Key Changes
1. **Unify expert map shape calculation logic**
- Ensure the shape of `expert_placement_map` and `log2phy_expert_map`
strictly aligns with the total number of experts (including redundant
experts) during initialization.
- Update the shape adjustment logic in EPLB dynamic update process to
match the initial expert map dimensions.

2. **Add shape consistency checks**
- Add assertion statements to verify the shape consistency of the two
maps after initialization and EPLB adjustment, preventing silent shape
mismatches in subsequent operations.

#### Impact
- Resolves tensor shape errors when using redundant experts with EPLB on
Ascend platform.
- Ensures correct expert routing and load balancing for MoE models with
redundant expert configurations.
- No breaking changes to existing functionality; compatible with
non-redundant expert deployments.

- vLLM version: release/v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/ad32e3e19ccf0526cb6744a5fed09a138a5fb2f9

---------

Signed-off-by: Che Ruan <cr623@ic.ac.uk>
Signed-off-by: shenchuxiaofugui <1311027364@qq.com>
Co-authored-by: Che Ruan <cr623@ic.ac.uk>
Co-authored-by: shenchuxiaofugui <1311027364@qq.com>
2026-01-06 17:22:36 +08:00
Zetong Li fe3f2c7702 [Refactor][EAGLE] 3/N delete redundant methods in mtp_proposer (#5420)
### What this PR does / why we need it?
This PR aims to delete redundant methods in mtp_proposer. All the
deleted methods now can be found in eagle_proposer. We also remove some
methods in eagle_proposer since they are identical to those in
vllm-eagle.

### Does this PR introduce _any_ user-facing change?
N/A

### How was this patch tested?
by ci

- vLLM version: release/v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/81786c87748b0177111dfdc07af5351d8389baa1

---------

Signed-off-by: Zetong Li <slippersss@126.com>
2026-01-06 16:47:39 +08:00
Shanshan Shen b94d589769 [MM][Bugfix] Update hf_config to hf_text_config (#5319)
### What this PR does / why we need it?

Following https://github.com/vllm-project/vllm-ascend/pull/5205, update
`hf_config` to `hf_text_config`.

Find more details at
https://github.com/vllm-project/vllm-ascend/pull/5205#issuecomment-3675417534
and
https://github.com/vllm-project/vllm-ascend/pull/5205#issuecomment-3677920872.

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: release/v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/5fbfa8d9ef15948599631baeb91e8220b2ee9bcc

Signed-off-by: shen-shanshan <467638484@qq.com>
2026-01-06 16:41:39 +08:00
Magnus 293b2275df [CI] Specify the version of xlite (#5612)
### What this PR does / why we need it?
Pin the xlite version to avoid CI failures during its upgrade.

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: changdawei1 <changdawei3@huawei.com>
2026-01-06 16:02:16 +08:00
wjunLu b8f245792e [Main2Main] Upgrade vllm commit to 0106 (#5617)
### What this PR does / why we need it?
Upgrade vllm commit to 0106

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/8be6432bdaf6275664d857b1e5e9bf8ed1ce299e

Signed-off-by: wjunLu <wjunlu217@gmail.com>
2026-01-06 15:50:40 +08:00
meihanc c1dcddce3f [CI]update bisheng version (#5621)
### What this PR does / why we need it?
update bisheng version in 20260105

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/8be6432bdaf6275664d857b1e5e9bf8ed1ce299e

Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
2026-01-06 15:22:22 +08:00
Qiu e07938047e [UT][PCP&DCP] UT for block_table.py (#5032)
## Purpose
This PR add unit test for `compute_slot_mapping` function in
`block_table.py` with various `pcp_size` & `dcp_size` &
`cp_kv_cache_interleave_size`.

## Test Plan
```
pytest tests/ut/worker/test_block_table.py
```
## Test Result
```
==== 3 passed, 2 warnings in 0.20s ====
```
- vLLM version: v0.12.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/ad32e3e19ccf0526cb6744a5fed09a138a5fb2f9

---------

Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
2026-01-06 11:19:25 +08:00
wjunLu 3cf059a72b [Main2Main] Upgrade vllm commit to 0105 (#5595)
### What this PR does / why we need it?

Upgrade vllm commit to 0105 (8be6432bdaf6275664d857b1e5e9bf8ed1ce299e)

1. Remove `maybe_padded_num_tokens` arg in `model_runner_v1.py` since
https://github.com/vllm-project/vllm/pull/31517 deleted unused arg

2. Remove dense `Qwen/Qwen3-0.6B` in
`tests/e2e/multicard/test_aclgraph_capture_replay.py` and
`tests/e2e/multicard/test_data_parallel.py` due to
https://github.com/vllm-project/vllm/pull/30739
where offline data parallel mode will not be supported/useful for dense
models

3. Adapt `vllm_ascend/worker/worker.py` due to
https://github.com/vllm-project/vllm/pull/31584

4. Adapt `self.block_size` calling due to
https://github.com/vllm-project/vllm/pull/31540

5. Modify `test_mla_v1.py` due to
https://github.com/vllm-project/vllm/pull/28454 , which refactorred
`get_head_size()`

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: wjunLu <wjunlu217@gmail.com>
2026-01-06 08:44:29 +08:00
Li Wang c5e2f48510 [CI] mv ops to correct path (#5615)
### What this PR does / why we need it?
mv ops to correct path
:`tests/e2e/nightly/single_node/ops/singlecard_ops/triton`

Signed-off-by: wangli <wangli858794774@gmail.com>
2026-01-05 23:17:07 +08:00
dsxsteven 129ba9fe1b [BugFix] Fix Smoke Testing Bug for DSR1 longseq (#5613)
### What this PR does / why we need it?
Fix Smoke Testing Bug for DSR1 longseq
We need to make this change because the daily smoke test case is
throwing an error: "max_tokens or max_completion_tokens is too large:
32768.This model's maximum context length is 32768 tokens and your
request has 128 input tokens". We encounter this error due to
max-out-len equals to max-model-len. We can fix this error by increasing
max-model-len argument in the script.
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?


- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: daishixun <dsxsteven@sina.com>
2026-01-05 22:40:28 +08:00
ZixuanWang 8eae949d11 Revert "[Feat] enable hierarchical mc2 ops on A2 by default (#5545)" (#5611)
This reverts commit fb9fdcdbe4.

### What this PR does / why we need it?
this pr breaks the smoke test because of that leads the error of
aclnnNeScalar:Kernel Run failed. opType: 25, NotEqual
        launch failed for NotEqual, errno:361001
<img width="1149" height="166"
alt="A6C9453D-4F0B-4256-DD80-A9C181DAB2D9"
src="https://github.com/user-attachments/assets/cab9c4b8-3fd1-4c6b-b424-474b46042726"
/>

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: zxwang <1476209578@qq.com>
2026-01-05 22:39:05 +08:00
Angazenn 11e75494b1 [TRITON][TEST]Add nightly test for triton split_qkv_rmsnorm_rope (#5267)
### What this PR does / why we need it?
Add nightly test for triton split_rmsnorm_rope

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: release/v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/ad32e3e19ccf0526cb6744a5fed09a138a5fb2f9

---------

Signed-off-by: Angazenn <supperccell@163.com>
2026-01-05 21:35:37 +08:00
Chen Chen a2daacbd71 [perf] Fix MLAPO weight disposal for KV-consumer MLA in PD-mix deploy... (#5192)
### What this PR does / why we need it?

- Problem: In MLA+MLAPO, KV-consumer deployments keep
fused_qkv_a_proj/q_proj weights and quant params even though MLAPO uses
the prepacked buffers, increasing memory footprint on decode nodes.
- Fix: Conditionally drop those tensors only when
`kv_transfer_config.is_kv_consumer` to reclaim memory (consistent with
the SFA behavior #4774 ).

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.12.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/ad32e3e19ccf0526cb6744a5fed09a138a5fb2f9

Signed-off-by: Chen Chen <0109chenchen@gmail.com>
2026-01-05 21:29:45 +08:00
Qiu b10ef9b9f3 [docs] Correct image about prefill phase of PCP (#5598)
### What this PR does / why we need it?
Remove the incorrectly depicted DCP all_gather operation in the prefill
stage PCP for GQA diagram.

Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
2026-01-05 20:21:59 +08:00
meihanc a034941d06 [CI] update triton-ascend version (#5584)
### What this PR does / why we need it?
update triton-ascend version to 20260105

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

---------

Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
2026-01-05 20:20:11 +08:00
Chao Lei 473431e7e2 [P/D]Remove mooncake kvpool unused parameter local_hostname (#5574)
### What this PR does / why we need it?
In mooncake kvpool, `local_hostname` is not used. Instead, the local IP
is obtained directly via `get_ip()`. Therefore, remove this parameter to
avoid confusion.

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/7157596103666ee7ccb7008acee8bff8a8ff1731

Signed-off-by: LCAIZJ <leichao139636@163.com>
2026-01-05 20:18:59 +08:00
Debonet d86021f7b4 [Bugfix] record cos and sin cache in AscendRotaryEmbedding (#5516)
### What this PR does / why we need it?

In scenarios where models like
[Moonlight](https://modelscope.cn/models/moonshotai/Moonlight-16B-A3B-Instruct)
(using MLA but without `rope_scaling` in config.json) invoke
`AscendRotaryEmbedding`. `_cos_cache` and `_sin_cache` are not recorded
correctly.

### Does this PR introduce _any_ user-facing change?

No

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/45c1ca1ca1ee8fa06df263c8715e8a412ff408d4

Signed-off-by: Debonex <719893090@qq.com>
2026-01-05 20:12:41 +08:00
meihanc 16b1bee804 [bugfix] fix test_camem failed with triton-ascend (#5492)
### What this PR does / why we need it?
This fixes a bug that occurred when running `test_camem.py` in the
triton-ascend environment `NPU function error:
aclrtGetMemInfo(ACL_HBM_MEM, &device_free, &device_total)`

- vLLM version: v0.13.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/5326c89803566a131c928f7fdd2100b75c981a42

---------

Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
2026-01-05 20:10:54 +08:00