## Summary
- replace the MiniMax usage accounting monkey patch with a runtime
wrapper implementation instead of source-text rewriting
- preserve MiniMax reasoning-token semantics when `</think>` is missing
by counting the emitted output as reasoning tokens
- add unit coverage for usage tracking helpers and MiniMax
reasoning-token counting
## Why
The previous implementation rewrote `OpenAIServingChat` by matching
exact source blocks. That was brittle against `vllm` source drift and
could crash during early plugin initialization with:
`RuntimeError: Failed to locate expected block while patching
OpenAIServingChat usage accounting.`
This change keeps the usage-accounting backport, but applies it by
wrapping the original stream/full generators and tracking output token
ids at runtime.
For MiniMax reasoning counting, a missing `</think>` should not be
treated as zero reasoning tokens. It can mean the whole output is still
in thinking mode, or that generation stopped before the closing token
was produced. In that case, the emitted output should still be counted
as reasoning.
## Validation
- `pytest -q
tests/ut/patch/platform/test_patch_minimax_usage_accounting.py`
- `vllm serve --help`
Signed-off-by: QwertyJack <7554089+QwertyJack@users.noreply.github.com>
Co-authored-by: QwertyJack <7554089+QwertyJack@users.noreply.github.com>
### What this PR does / why we need it?
This rebases the GLM47 tool-call parser fix onto `releases/v0.18.0`
after the MiniMax usage-accounting patch merged upstream on March 27,
2026.
It fixes OpenAI chat tool-call streaming for GLM47 by:
- draining terminal parser chunks that contain both the final argument
text and the closing `</tool_call>` suffix
- computing finish backfill from the tool argument bytes actually
emitted to the client, instead of trusting parser-internal buffered
state
- adding focused regression tests for finish backfill and terminal chunk
handling
### Does this PR introduce _any_ user-facing change?
Yes. GLM47 OpenAI-compatible streaming tool-call responses now emit
correct final chunks and argument payloads on `releases/v0.18.0`.
### How was this patch tested?
- `pytest -q tests/ut/patch/platform/test_patch_glm_tool_call_parser.py
tests/ut/patch/platform/test_patch_minimax_usage_accounting.py`
- `python -m pre_commit run --files
vllm_ascend/patch/platform/patch_glm_tool_call_parser.py
tests/ut/patch/platform/test_patch_glm_tool_call_parser.py
vllm_ascend/patch/platform/__init__.py vllm_ascend/patch/__init__.py`
---------
Signed-off-by: QwertyJack <7554089+QwertyJack@users.noreply.github.com>
Co-authored-by: QwertyJack <7554089+QwertyJack@users.noreply.github.com>
### What this PR does / why we need it?
This backports the MiniMax M2 reasoning-token usage accounting fix onto
`releases/v0.18.0` for vllm-ascend.
The release branch does not include the other local GLM patch commit, so
this PR keeps the MiniMax change self-contained by:
- registering `patch_minimax_usage_accounting` on the release branch
- backporting `completion_tokens_details.reasoning_tokens` into chat
usage generation
- fixing MiniMax reasoning token counting for `</think>`-delimited
outputs without depending on the GLM suffix patch
### Does this PR introduce _any_ user-facing change?
Yes. OpenAI-compatible chat usage accounting for MiniMax M2 responses
now reports corrected reasoning token counts on the release branch.
### How was this patch tested?
- `python -m compileall
vllm_ascend/patch/platform/patch_minimax_usage_accounting.py`
- `python - <<'PY'` import check for
`vllm_ascend.patch.platform.patch_minimax_usage_accounting` on top of
`releases/v0.18.0`
No targeted automated regression test exists for this release-branch
backport yet, so I validated syntax and module import compatibility on
the release branch.
---------
Signed-off-by: QwertyJack <7554089+QwertyJack@users.noreply.github.com>
Co-authored-by: QwertyJack <7554089+QwertyJack@users.noreply.github.com>
### What this PR does / why we need it?
This PR introduces a "balance scheduling" feature, enabled by the
`VLLM_ASCEND_BALANCE_SCHEDULING` environment variable. This feature
adjusts the scheduling logic to better balance the load across
data-parallel workers, preventing a single worker from blocking
scheduling for others. This can improve overall throughput.
Additionally, this PR includes a number of other updates and fixes to
the scheduler, syncing it with a more recent version of the upstream
vLLM scheduler. These changes include:
- Handling for paused scheduler state.
- Support for Mamba block-aligned splits.
- Handling for streaming requests.
- Refinements in preemption logic and resource management (KV cache,
encoder cache).
- General code refactoring for clarity and correctness.
Fixes #
### Does this PR introduce _any_ user-facing change?
Yes, this PR introduces a new feature controlled by the
`VLLM_ASCEND_BALANCE_SCHEDULING` environment variable. When enabled, the
scheduling behavior changes, which could affect performance and request
throughput.
### How was this patch tested?
CI passed. Further testing should be done to validate the performance
and correctness of the new scheduling logic under various workloads,
with and without the feature flag enabled.
Signed-off-by: GDzhu01 <809721801@qq.com>
### What this PR does / why we need it?
1. upgrade to 0.18.0
2. ensure kernel_block_sizes is int for Eagle drafter
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
- vLLM version: v0.17.0
- vLLM main:
8b6325758c
---------
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: hfadzxy <starmoon_zhang@163.com>
Co-authored-by: hfadzxy <starmoon_zhang@163.com>
### What this PR does / why we need it?
1. Mamba Cache Support on 310P: Implemented logic to correctly
initialize and allocate KV cache for Mamba models on the 310P platform,
including handling of state tensors and page size alignment.
2. Increased Attention Head Size Support: Modified the attention backend
to support attn_head_size larger than 128 by dynamically selecting
appropriate kernel block sizes based on hardware limitations (e.g.,
block_size * head_size <= 16384).
3. Refactored KV Cache Allocation: Consolidated and improved the KV
cache allocation mechanism, moving from separate size calculation and
allocation steps to a unified _allocate_kv_cache_tensors method that
handles both Attention and Mamba specific cache structures.
4. Dynamic Mamba Config Patching: Introduced conditional loading of
Mamba configuration patches, specifically using patch_mamba_config_310
for the 310P platform to ensure platform-specific optimizations and
validations.
5. Reserve reasonable memory to allocate KV cache to avoid OOM issue
with default gpu_memory_utilization.
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
Qwen3.5 E2E test
- vLLM version: v0.17.0
- vLLM main:
4034c3d32e
---------
Signed-off-by: pu-zhe <zpuaa@outlook.com>
### What this PR does / why we need it?
1.fix "TypeError: get_attn_backend() remove variable": [Refactor
`check_and_update_config`](https://github.com/vllm-project/vllm/pull/35122)
2.fix [Rename `compile_ranges_split_points` to
`compile_ranges_endpoints`](https://github.com/vllm-project/vllm/pull/36027)
3.fix "RuntimeError: device_allocator not a DeviceAllocator":[Replace
memory related torch.cuda
APIs"](https://github.com/vllm-project/vllm/pull/37031)
4.fix [Support multiple KV groups in OffloadingSpec
](https://github.com/vllm-project/vllm/pull/36610) removed
self.offloaded_block_size and changed self.gpu_block_size from a scalar
to a tuple of per-group block sizes, adding block_size_factor.
5.fix [Consolidate
SupportsEagle](https://github.com/vllm-project/vllm/pull/36063) renamed
get_eagle3_aux_hidden_state_layers() to
get_eagle3_default_aux_hidden_state_layers() and added a
supports_eagle3() guard before calling it.
### Does this PR introduce _any_ user-facing change?
NA
### How was this patch tested?
E2E
- vLLM version: v0.17.0
- vLLM main:
8a680463fa
---------
Signed-off-by: leo-pony <nengjunma@outlook.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
### What this PR does / why we need it?
This PR fixes the logger initialization in patches so that the log info
can be displayed as expected.
### Does this PR introduce _any_ user-facing change?
No.
- vLLM version: v0.17.0
- vLLM main:
4497431df6
---------
Signed-off-by: Angazenn <supperccell@163.com>
### What this PR does / why we need it?
This PR restores #7029, which adds W8A8C8 support for dsv3.2/glm5 using
the `lightning_indexer_quant` ops in the pd-mix stage.
The original PR was reverted by #7288 because the patch did not work
with the recompute scheduler.
This PR also fixes the patching issue so that it works correctly with
the recompute scheduler.
### Does this PR introduce _any_ user-facing change?
Yes. To enable LI C8, users need to set the `enable_sparse_c8` option to
`"true"` in `additional_config`.
- vLLM version: v0.17.0
- vLLM main:
4034c3d32e
---------
Signed-off-by: rjg-lyh <1318825571@qq.com>
### What this PR does / why we need it?
This reverts commit 7ed9e9de69, which
introduces an issue that the patch doesn't work with recompute scheduler
enabled.
- vLLM version: v0.17.0
- vLLM main:
4034c3d32e
---------
Signed-off-by: MengqingCao <cmq0113@163.com>
### What this PR does / why we need it?
This PR supports W8A8C8 in dsv3.2/glm5 with lightning_indexer_quant ops
in pd-mix stage mainly.
Because the code for the current PD-disaggregated scenario is still
under refactoring and cleanup, this PR prioritizes ensuring the C8
functionality in the pd-mix scenario.
The next steps are planned in two parts:
① Once the optimized scatter operator is updated, we will replace the
original operator to improve the performance of storing k_scale.
② Once the code logic for the PD-disaggregated scenario becomes stable,
we will carry out more comprehensive validation and make appropriate
adaptations.
③ Because enabling C8 currently introduces several new operators whose
performance still needs improvement, performance may regress in some
scenarios. Therefore, only after all the operators are fully ready can
we ensure that this feature does not cause any performance degradation.
At that point, we will enable this feature by default and remove the
switch in `additional_config`.
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
CI passed with new added/existing test.
- vLLM version: v0.16.0
- vLLM main:
4034c3d32e
---------
Signed-off-by: rjg-lyh <1318825571@qq.com>
### What this PR does / why we need it?
Initial version to support minimax-m2.5 on vllm-ascend.
This commit coverting original fp8 weight to a quantilized bf16 to
support Minimax-m2.5 on NPU.
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
- vLLM version: v0.16.0
- vLLM main:
4034c3d32e
### Test Report
Self tested precision summary, where the official precision score of
AIME2025 is 86.3
<img width="426" height="84" alt="image"
src="https://github.com/user-attachments/assets/a3ce2452-92fa-4713-962e-862248e0b61a"
/>
---------
Signed-off-by: limuyuan <limuyuan3@huawei.com>
Signed-off-by: SparrowMu <52023119+SparrowMu@users.noreply.github.com>
Co-authored-by: limuyuan <limuyuan3@huawei.com>
### What this PR does / why we need it?
Supports contiguous tensor hybrid-attn kv-cache on fullattn-mamba hybrid
model, such as Qwen3Next and Qwen3.5.
Due to the restrictions of Ascend operators, all KV tensors, conv
tensors, and SSM tensors must be contiguous. Therefore, this PR uses the
following solution to generate the KV cache:
tensor1: [(kv_padding), conv , ...]
tensor2: [k , ssm , ...]
tensor3: [v , (mamba_padding), ...]
Under this scheme, although some waste may occur, the tensors of all
caches are guaranteed to be contiguous.
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
By CI.
- vLLM version: v0.16.0
- vLLM main:
15d76f74e2
---------
Signed-off-by: nwpu-zxr <zhouxuerong2@huawei.com>
### What this PR does / why we need it?
This PR upgrades the vLLM dependency from `v0.14.1` to `v0.15.0`. This
involves:
- Updating the `VLLM_TAG` in all `Dockerfile`.
- Updating the vLLM version in `docs/source/conf.py`.
- Removing conditional code paths specific to `v0.14.1` across the
codebase, which simplifies maintenance.
- Fix `TypeError: MMEncoderAttention.__init__() got an unexpected
keyword argument 'multimodal_config'` due to
https://github.com/vllm-project/vllm/pull/31972.
- Fix `_shared_experts: 'NoneType' object is not callable` due to
https://github.com/vllm-project/vllm/pull/32082 by
https://github.com/vllm-project/vllm-ascend/pull/6335.
- Fix `ReshapeAndCacheOperation setup failed!` due to
https://github.com/vllm-project/vllm/pull/25954 by overriding attention
metadata slots.
This upgrade is necessary to keep the project aligned with the latest
features, bug fixes, and API changes in the vLLM project.
### Does this PR introduce _any_ user-facing change?
No, this is an internal dependency update and does not introduce any
user-facing changes.
### How was this patch tested?
CI is expected to pass with these changes, ensuring that all existing
tests are successful with the new vLLM version.
- vLLM version: v0.14.1
- vLLM main:
dc917cceb8
co-authored-by: shen-shanshan <467638484@qq.com>
---------
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
### What this PR does / why we need it?
Drop vLLM 0.13.0 support, upgrade to 0.14.0
- vLLM version: v0.13.0
- vLLM main:
d68209402d
---------
Signed-off-by: hfadzxy <starmoon_zhang@163.com>
### What this PR does / why we need it?
Add basic 310p support. Only dense models work with eager mode now.
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
---------
Signed-off-by: Tflowers-0129 <2906339855@qq.com>
Signed-off-by: Shaoxu Cheng <2906339855@qq.com>
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:
2f4e6548ef
---------
Signed-off-by: wxsIcey <1790571317@qq.com>
### 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>
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: release/v0.13.0
- vLLM main:
ad32e3e19c
---------
Signed-off-by: wxsIcey <1790571317@qq.com>
### What this PR does / why we need it?
Fix the bug in the PCP overlay feature
1、Fix the bug related to PCP and EPLB overlap by including PCP size in
the word_size calculation.
2、In the PCP pooling scenario, a prompt has been added for setting the
cp_kv_cache_interleave_size.
- vLLM version: v0.13.0
- vLLM main:
7157596103
Signed-off-by: weiguihua2 <weiguihua2@huawei.com>
Currently, the vllm pr: https://github.com/vllm-project/vllm/pull/24252
is causing operator fusion to fail, which can be mitigated by patching
the backend. Once the problem is completely resolved, I will submit a
new pull request to remove the patch.
- vLLM version: release/v0.13.0
- vLLM main:
5fbfa8d9ef
---------
Signed-off-by: wxsIcey <1790571317@qq.com>
### Motivation.
**Limitations of the current vLLM v1 scheduling strategy**
vLLM v1 scheduling currently enables chunkedprefill by default, which
processes prefill and decode requests simultaneously in a single
scheduling session. This can impact the overall system throughput and
performance in some scenarios.
Balance scheduling addresses this issue by synchronizing the number of
running queues across all schedulers to delay the scheduling of new
requests, thereby improving the overall system's steady-state decoding
time. This achieves:
✅Adding `balance_gather` to the scheduler synchronizes the number of
requests in the running queues between DPs.
✅Balance scheduling improves the decode steady-state time, thereby
increasing the overall output throughput of the inference system.
### Proposed Change.
**1.Feature Overview**
In the vLLM scheduler, running requests (i.e., requests that are already
undergoing pre-filled computation) have the highest priority, followed
by waiting requests (i.e., requests that have not yet been computed).
As shown in the diagram above, when the entire inference system exits
from a steady state, the scheduler will schedule a batch of new requests
for prefill operations and then synchronize them among the dynamic
programming (DP) models. This can cause some DP models that are entirely
decoded to synchronize with the number of prefilled tokens. Frequent
prefill scheduling by certain DP models can lead to a deterioration in
the overall system output throughput.
Balance scheduling synchronizes the number of running queue requests
across different DPs, and only schedules new requests for prefilling
when at least every scheduler has fewer than max_nun_requst.
**2.Implementation Design**
**3.Experiment Results**
- Fixed-length input scenario: In the performance test scenario with
3.5K fixed-length input and 1.5K fixed-length output, the throughput
performance was improved by approximately **18%** after adding balance
scheduling.
| Method | Model | Input Len | Request Count | Output Len | BatchSize |
Average TTFT | Average TPOT | e2e duration | Input Token Throughput |
Output Token Throughput | Request Throughput
| ---- | ---- | ---- | ---- | ---- | ---- | ---- | ---- | ---- | ---- |
---- | ---- |
| Baseline | DeepSeekV3.1 | 3500 | 512 | 1500 | 128 | 6600 | 86.85 |
591.9s | 3030.5 | 1297.3 | 0.86 |
| Balance scheduling | DeepSeekV3.1 | 3500 | 512 | 1500 | 128 | 7012 |
70.63 | 501.7s | 3575.7 | 1530.7 | 1.02 |
**4.Demo PR**
[#29721 ](https://github.com/vllm-project/vllm/pull/29721)
---------
Signed-off-by: GDzhu01 <809721801@qq.com>
We decided to release v0.13.0 soon. So no need to support 0.12.0 now.
Let's drop it.
- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
### What this PR does / why we need it?
This PR integrate suffix decoding (https://arxiv.org/abs/2411.04975)
from vllm (https://github.com/vllm-project/vllm/pull/25784)
#
Suffix Decoding is a dynamic n-gram matching method that:
1. Uses suffix trees to generate speculative tokens quickly using branch
frequency counts.
2. Can keep a history of prior model responses, which tends to work very
well with repetitive agentic use cases.
3. Can be dynamically updated with newly generated tokens, and FIFO
eviction of older requests.
#
### Does this PR introduce _any_ user-facing change?
This feature should be implemented as opt-in and remain seamless for
users who do not require suffix speculative decoding.
For users who wish to enable it, they must first install
arctic-inference:
`pip install arctic-inference
`
After installation, the suffix speculative decoding feature can be
enabled using the following speculative config:
`--speculative_config '{"method": "suffix", "num_speculative_tokens":
5}'
`
### How was this patch tested?
This PR is currently being tested on vLLM
main:83f478bb19
with PR https://github.com/vllm-project/vllm/pull/25784
In our previous testing, suffix decoding achieved a 13%-30% throughput
improvement over n-gram on the sonnet dataset, tested on vllm-ascend
v0.9.1 with concurrency ranging from 2 to 40.
- vLLM version: v0.11.2
---------
Signed-off-by: fluctlux <38945811+fluctlux@users.noreply.github.com>
### What this PR does / why we need it?
Currently, there are two paths to judge the chip type in code,
`get_ascend_soc_version` use `get_soc_version` api in torch_npu, and
`is_310p` `use _build_info.__soc_version__`, which generate when
install. We need to unify the two paths.
We need to unify these codes based on the following points:
1. We need to ensure consistency in chip type judgment between compiling
and running states;
2. In compiling state, we need chip type to complete op's compilation,
but in running state, we only need device
type(910B/910_93/310P/910_95/etc) to make code branch judgement;
3. In compiling state, torch_npu may not have been installed yet, so we
can't use torch_npu's api.
Based on the above points, we have made the following changes:
1. When user set env `SOC_VERSION`, use it; when not set, query
soc_version by `npu-smi`;
2. generate device_type based on soc_version when compiling, and write
`__device_type__` instead of `__soc_version__` in `_build_info.py`;
3. In running state, use `__device_type__` to judge code branch.
### Does this PR introduce _any_ user-facing change?
When not set env `SOC_VERSION`, it will not be `ASCEND910B1` by default,
we will query soc_version by `npu-smi`. And env `SOC_VERSION` must be in
the list `soc_to_device` in `setup.py`.
- vLLM version: v0.11.0
- vLLM main:
2918c1b49c
Signed-off-by: zzzzwwjj <1183291235@qq.com>
### What this PR does / why we need it?
Fix a bug caused by this pr:
https://github.com/vllm-project/vllm-ascend/pull/4223
The bug makes
vllm-ascend/vllm_ascend/patch/platform/patch_multiproc_executor.py patch
in a wrong way
### How was this patch tested?
Tested in a single node. When the environment DYNAMIC_EPLB is set to
true, the patch works correctly. When it's set to false, the patch do
not patch
- vLLM version: v0.11.0
- vLLM main:
2918c1b49c
Signed-off-by: 白永斌 <baiyongbin3@h-partners.com>
Co-authored-by: 白永斌 <baiyongbin3@h-partners.com>
There is a lot hack code for v0.11.0, which makes the code hard to
upgrade to newer vLLM version. Since v0.11.0 will release soon. Let's
drop v0.11.0 support first. Then we'll upgrade to v0.11.2 soon.
- vLLM version: v0.11.0
- vLLM main:
2918c1b49c
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
Drop VLLM_USE_V1 usage. This env has been removed from vLLM already.
- vLLM version: v0.11.0
- vLLM main:
83f478bb19
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
### What this PR does / why we need it?
Fixes a compatible bug with `torch_npu.npu_fused_infer_attention_score`
which is discribed in
https://github.com/vllm-project/vllm-ascend/issues/4020.
@momo609 tells us this solution.
### Does this PR introduce _any_ user-facing change?
N/A
### How was this patch tested?
The environment is same with this issue,
https://github.com/vllm-project/vllm-ascend/issues/4020.
We modify the code according to
https://github.com/vllm-project/vllm-ascend/pull/3918.
And run below codes:
```python
# run with Qwen3-next-mtp
prompts = [
"Who are you?",
]
sampling_params = SamplingParams(temperature=0.0, top_p=0.95, top_k=40, max_tokens=128)
llm = LLM(model="/home/model/Qwen3-Next-80B-A3B-Instruct",
tensor_parallel_size=4,
enforce_eager=True,
distributed_executor_backend="mp",
gpu_memory_utilization=0.7,
speculative_config={
"method": "qwen3_next_mtp",
"num_speculative_tokens": 1,
},
max_model_len=4096)
outputs = llm.generate(prompts, sampling_params)
for output in outputs:
prompt = output.prompt
generated_text = output.outputs[0].text
print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
```
Outputs:
```text
Prompt: 'Who are you?', Generated text: ' I am Qwen, a large-scale language model independently developed by the Tongyi Lab under Alibaba Group. I am designed to answer questions, create text such as stories, official documents, emails, scripts, and more, as well as perform logical reasoning, programming, and other tasks. If you have any questions or need assistance, feel free to let me know anytime!'
```
Now, `torch_npu.npu_fused_infer_attention_score` is compatible with
Qwen3-Next.
- vLLM version: v0.11.0
- vLLM main:
83f478bb19
Signed-off-by: drslark <slarksblood@qq.com>
