drslark 0fb1dc43a1 [BugFix][main] Adapted Qwen3-Next-MTP to chunked prefill (#4770) ...

### What this PR does / why we need it?
The pad `-1` modification is from
https://github.com/vllm-project/vllm/pull/25743.

It still has bugs for batched chunked prefill.

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

Signed-off-by: drslark <slarksblood@qq.com>
Co-authored-by: Mengqing Cao <cmq0113@163.com>

2025-12-10 22:54:24 +08:00

..

test_aclgraph_capture_replay.py

[Fix] Add extra warmup run count for MC2 on specific SoC version (#4843)

2025-12-09 21:37:38 +08:00

test_chunk_gated_delta_rule.py

【OPS】qwen3-next support triton chunk_gated_delta_rule ops (#4070)

2025-11-28 20:55:43 +08:00

test_data_parallel_tp2.py

upgrade torch npu version (#4433)

2025-12-01 19:01:55 +08:00

test_data_parallel.py

[Test] Temporarily skips Qwen3-30B-A3B-W8A8 data parallel test case (#4857)

2025-12-10 11:05:32 +08:00

test_expert_parallel.py

[CI] drop ascend scheduler test (#4582)

2025-12-01 20:33:50 +08:00

test_external_launcher.py

upgrade torch npu version (#4433)

2025-12-01 19:01:55 +08:00

test_full_graph_mode.py

support FULL graph mode for GQA (#3970)

2025-11-17 10:50:35 +08:00

test_fused_moe_allgather_ep.py

[CI] drop ascend scheduler test (#4582)

2025-12-01 20:33:50 +08:00

test_ilama_lora_tp2.py

ACLgraph enable: Test cases revisions for all features (#3388)

2025-10-17 17:15:19 +08:00

test_offline_inference_distributed.py

[Feat] Support native Kimi-K2-Thinking native W4A16 quantized experts weights (#4516)

2025-12-10 15:58:52 +08:00

test_offline_weight_load.py

[MOE]move weight transpose to wakeup for RL secnarios (#4626)

2025-12-08 20:34:52 +08:00

test_pipeline_parallel.py

[bugfix] fix pipeline parallel for mla & sfa attention backend (#3459)

2025-10-15 17:13:27 +08:00

test_prefix_caching.py

[CI]enable chunked prefill by default (#4569)

2025-12-02 08:54:34 +08:00

test_quantization.py

[Quantization] Support compressed tensors w8a8 static and w8a8 dynamic weight (#4036)

2025-11-28 14:09:39 +08:00

test_qwen3_moe.py

Reapply "[MoE] [Refactor] Remove manual memory cleanup (#3365)" (#3483) (#3512)

2025-10-22 11:41:30 +08:00

test_qwen3_next.py

[BugFix][main] Adapted Qwen3-Next-MTP to chunked prefill (#4770)

2025-12-10 22:54:24 +08:00

test_shared_expert_dp.py

[Feat] shared expert dp for deepseek_mtp (#3811)

2025-12-01 20:44:11 +08:00

test_single_request_aclgraph.py

Drop 0.11.0 support (#4377)

2025-11-24 17:08:20 +08:00

test_weight_loader.py

upgrade torch npu version (#4433)

2025-12-01 19:01:55 +08:00