biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Model] Enable LoRA support for tower and connector in DotsOCR (#31825)

Browse Source 
Signed-off-by: ShaanveerS <shaanver.singh@gmail.com>
This commit is contained in:
ShaanveerS
2026-01-08 07:50:16 +01:00
committed by GitHub
parent 5f2a473ff3
commit 9572f74f15
2 changed files with 9 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/models/supported_models.md
View File

@@ -371,7 +371,7 @@ th {
| `DeepseekV2ForCausalLM` | DeepSeek-V2 | `deepseek-ai/DeepSeek-V2`, `deepseek-ai/DeepSeek-V2-Chat`, etc. | ✅︎ | ✅︎ | | `DeepseekV2ForCausalLM` | DeepSeek-V2 | `deepseek-ai/DeepSeek-V2`, `deepseek-ai/DeepSeek-V2-Chat`, etc. | ✅︎ | ✅︎ |
| `DeepseekV3ForCausalLM` | DeepSeek-V3 | `deepseek-ai/DeepSeek-V3`, `deepseek-ai/DeepSeek-R1`, `deepseek-ai/DeepSeek-V3.1`, etc. | ✅︎ | ✅︎ | | `DeepseekV3ForCausalLM` | DeepSeek-V3 | `deepseek-ai/DeepSeek-V3`, `deepseek-ai/DeepSeek-R1`, `deepseek-ai/DeepSeek-V3.1`, etc. | ✅︎ | ✅︎ |
| `Dots1ForCausalLM` | dots.llm1 | `rednote-hilab/dots.llm1.base`, `rednote-hilab/dots.llm1.inst`, etc. | | ✅︎ | | `Dots1ForCausalLM` | dots.llm1 | `rednote-hilab/dots.llm1.base`, `rednote-hilab/dots.llm1.inst`, etc. | | ✅︎ |
| `DotsOCRForCausalLM` | dots_ocr | `rednote-hilab/dots.ocr` | | ✅︎ | | `DotsOCRForCausalLM` | dots_ocr | `rednote-hilab/dots.ocr` | ✅︎ | ✅︎ |
| `Ernie4_5ForCausalLM` | Ernie4.5 | `baidu/ERNIE-4.5-0.3B-PT`, etc. | ✅︎ | ✅︎ | | `Ernie4_5ForCausalLM` | Ernie4.5 | `baidu/ERNIE-4.5-0.3B-PT`, etc. | ✅︎ | ✅︎ |
| `Ernie4_5_MoeForCausalLM` | Ernie4.5MoE | `baidu/ERNIE-4.5-21B-A3B-PT`, `baidu/ERNIE-4.5-300B-A47B-PT`, etc. |✅︎| ✅︎ | | `Ernie4_5_MoeForCausalLM` | Ernie4.5MoE | `baidu/ERNIE-4.5-21B-A3B-PT`, `baidu/ERNIE-4.5-300B-A47B-PT`, etc. |✅︎| ✅︎ |
| `ExaoneForCausalLM` | EXAONE-3 | `LGAI-EXAONE/EXAONE-3.0-7.8B-Instruct`, etc. | ✅︎ | ✅︎ | | `ExaoneForCausalLM` | EXAONE-3 | `LGAI-EXAONE/EXAONE-3.0-7.8B-Instruct`, etc. | ✅︎ | ✅︎ |

8
vllm/model_executor/models/dots_ocr.py
View File

@@ -765,6 +765,14 @@ class DotsOCRForCausalLM(nn.Module, SupportsMultiModal, SupportsPP, SupportsLoRA
def get_language_model(self) -> torch.nn.Module: def get_language_model(self) -> torch.nn.Module:
return self.language_model return self.language_model
def get_num_mm_encoder_tokens(self, num_image_tokens: int) -> int:
merge_size = self.vision_tower.spatial_merge_size
return num_image_tokens * (merge_size**2)
def get_num_mm_connector_tokens(self, num_vision_tokens: int) -> int:
merge_size = self.vision_tower.spatial_merge_size
return num_vision_tokens // (merge_size**2)
def embed_multimodal(self, **kwargs: object) -> MultiModalEmbeddings: def embed_multimodal(self, **kwargs: object) -> MultiModalEmbeddings:
image_input = self._parse_and_validate_image_input(**kwargs) image_input = self._parse_and_validate_image_input(**kwargs)
if image_input is None: if image_input is None: