EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
Files
8813832387992ed69951f30cb872cca3bf240803
xc-llm-ascend/vllm_ascend/models/__init__.py

19 lines
764 B
Python
Raw Normal View History

register qwen2_vl to rewrite qwen2_vl forwad (#241) Add qwen2-vl ascend impletation. --------- Signed-off-by: zouyida <zouyida@huawei.com>
2025-03-07 15:41:47 +08:00
from vllm import ModelRegistry
def register_model():
[Model]Add support for qwen3_vl and qwen3_vl_moe (#3103) ### What this PR does / why we need it? This PR is for the adaptation and optimization of qwen3_vl and qwen3_vl_moe on the Ascend platform. ### Does this PR introduce _any_ user-facing change? None ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: https://github.com/vllm-project/vllm/commit/b1068903fdca26cf6b4a1a51a32c3365ce3ac636 --------- Signed-off-by: booker123456 <945658361@qq.com>
2025-09-25 18:50:12 +08:00
ModelRegistry.register_model(
"Qwen3VLMoeForConditionalGeneration",
[MM][Model][Perf] Remove Qwen2.5-VL modeling files and add patch for VisionAttention (#4349) ### What this PR does / why we need it? - [x] Patch `Qwen2_5_VisionAttention` with `AscendQwen2_5_VisionAttention`. - [x] Replace `AscendQwen2_5_VisionTransformer` with `Qwen2_5_VisionTransformer` in vllm. - [x] Move padding logic (q/k/v and cos/sin) before FA to `forward()` of `Qwen2_5_VisionAttention`. - [x] Covert `cu_seqlens` in `Qwen2_5_VisionAttention` from cumulative form to intervals and move it to cpu (compatible for npu FA). - [x] Remove Qwen2.5-VL modeling files. - [x] Remove Qwen2.5-VL (without padding) modeling files. - [x] Remove related UT. - [x] Make `set_forward_context` pluggable when getting MM embedding. Find more details at https://github.com/vllm-project/vllm/pull/29388. - [x] Simplify padding logic for FA. - [x] Add patch for https://github.com/vllm-project/vllm/pull/28798. ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? - [x] Functional test (eager mode) - [x] Functional test (graph mode) - [x] Benchmark - vLLM version: v0.11.2 --------- Signed-off-by: shen-shanshan <467638484@qq.com>
2025-11-28 14:23:00 +08:00
"vllm_ascend.models.qwen3_vl:AscendQwen3VLMoeForConditionalGeneration")
[Model]Add support for qwen3_vl and qwen3_vl_moe (#3103) ### What this PR does / why we need it? This PR is for the adaptation and optimization of qwen3_vl and qwen3_vl_moe on the Ascend platform. ### Does this PR introduce _any_ user-facing change? None ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: https://github.com/vllm-project/vllm/commit/b1068903fdca26cf6b4a1a51a32c3365ce3ac636 --------- Signed-off-by: booker123456 <945658361@qq.com>
2025-09-25 18:50:12 +08:00
ModelRegistry.register_model(
"Qwen3VLForConditionalGeneration",
[MM][Model][Perf] Remove Qwen2.5-VL modeling files and add patch for VisionAttention (#4349) ### What this PR does / why we need it? - [x] Patch `Qwen2_5_VisionAttention` with `AscendQwen2_5_VisionAttention`. - [x] Replace `AscendQwen2_5_VisionTransformer` with `Qwen2_5_VisionTransformer` in vllm. - [x] Move padding logic (q/k/v and cos/sin) before FA to `forward()` of `Qwen2_5_VisionAttention`. - [x] Covert `cu_seqlens` in `Qwen2_5_VisionAttention` from cumulative form to intervals and move it to cpu (compatible for npu FA). - [x] Remove Qwen2.5-VL modeling files. - [x] Remove Qwen2.5-VL (without padding) modeling files. - [x] Remove related UT. - [x] Make `set_forward_context` pluggable when getting MM embedding. Find more details at https://github.com/vllm-project/vllm/pull/29388. - [x] Simplify padding logic for FA. - [x] Add patch for https://github.com/vllm-project/vllm/pull/28798. ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? - [x] Functional test (eager mode) - [x] Functional test (graph mode) - [x] Benchmark - vLLM version: v0.11.2 --------- Signed-off-by: shen-shanshan <467638484@qq.com>
2025-11-28 14:23:00 +08:00
"vllm_ascend.models.qwen3_vl:AscendQwen3VLForConditionalGeneration")
[deepseek][bugfix] support deepseek quant (#469) - support deepseek quant - add w8a8_dynamic quant see #391 Signed-off-by: MengqingCao <cmq0113@163.com> Co-authored-by: zzzzwwjj <1183291235@qq.com>
2025-04-07 10:56:12 +08:00
[Misc] Remove pangu model file (#2798) vllm-ascend won't contain model file anymore. Now pangu model file has been moved to torchair module. The origin one can be removed. Note: After this PR, pangu only works with torchair mode then. - vLLM version: v0.10.1.1 - vLLM main: https://github.com/vllm-project/vllm/commit/8c892b1831a8a35fbf5a89fedfaae0ab9bef2dd0 Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
2025-09-08 21:30:37 +08:00
# There is no PanguProMoEForCausalLM in vLLM, so we should register it before vLLM config initialization
# to make sure the model can be loaded correctly. This register step can be removed once vLLM support PanguProMoEForCausalLM.
Support Pangu Pro MoE model (#1204) ### What this PR does / why we need it? Support Pangu Pro MoE model (https://arxiv.org/abs/2505.21411) ### Does this PR introduce _any_ user-facing change? Yes, new model supported ### How was this patch tested? Test locally --------- Signed-off-by: angazenn <zengyanjia@huawei.com> Co-authored-by: angazenn <zengyanjia@huawei.com>
2025-06-20 23:59:59 +08:00
ModelRegistry.register_model(
"PanguProMoEForCausalLM",
[Misc] Remove pangu model file (#2798) vllm-ascend won't contain model file anymore. Now pangu model file has been moved to torchair module. The origin one can be removed. Note: After this PR, pangu only works with torchair mode then. - vLLM version: v0.10.1.1 - vLLM main: https://github.com/vllm-project/vllm/commit/8c892b1831a8a35fbf5a89fedfaae0ab9bef2dd0 Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
2025-09-08 21:30:37 +08:00
"vllm_ascend.torchair.models.torchair_pangu_moe:PanguProMoEForCausalLM"
)
Reference in New Issue Copy Permalink