A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
[Model] Enable BLOOM on V1 (#23488)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Cyrus Leung committed
Aug 24, 2025 at 21:30 UTC
e2db1164a186a9d2592299aaa5aea3f013711db3
2 files changed
+3
-3
docs/models/supported_models.md
+1
-1
@@ -328,7 +328,7 @@ th {
328
| `BaiChuanForCausalLM` | Baichuan2, Baichuan | `baichuan-inc/Baichuan2-13B-Chat`, `baichuan-inc/Baichuan-7B`, etc. | ✅︎ | ✅︎ | ✅︎ |
329
| `BailingMoeForCausalLM` | Ling | `inclusionAI/Ling-lite-1.5`, `inclusionAI/Ling-plus`, etc. | ✅︎ | ✅︎ | ✅︎ |
330
| `BambaForCausalLM` | Bamba | `ibm-ai-platform/Bamba-9B-fp8`, `ibm-ai-platform/Bamba-9B` | ✅︎ | ✅︎ | ✅︎ |
331
-| `BloomForCausalLM` | BLOOM, BLOOMZ, BLOOMChat | `bigscience/bloom`, `bigscience/bloomz`, etc. | | ✅︎ | |
331
+| `BloomForCausalLM` | BLOOM, BLOOMZ, BLOOMChat | `bigscience/bloom`, `bigscience/bloomz`, etc. | | ✅︎ | ✅︎ |
332
| `BartForConditionalGeneration` | BART | `facebook/bart-base`, `facebook/bart-large-cnn`, etc. | | | |
333
| `MBartForConditionalGeneration` | mBART | `facebook/mbart-large-en-ro`, `facebook/mbart-large-50`, etc. | | | |
334
| `ChatGLMModel`, `ChatGLMForConditionalGeneration` | ChatGLM | `zai-org/chatglm2-6b`, `zai-org/chatglm3-6b`, `ShieldLM-6B-chatglm3`, etc. | ✅︎ | ✅︎ | ✅︎ |
vllm/model_executor/models/bloom.py
+2
-2
@@ -43,7 +43,7 @@ from vllm.model_executor.model_loader.weight_utils import default_weight_loader
43
from vllm.model_executor.sampling_metadata import SamplingMetadata
44
from vllm.sequence import IntermediateTensors
45
46
-from .interfaces import SupportsPP, SupportsQuant, SupportsV0Only
46
+from .interfaces import SupportsPP, SupportsQuant
47
from .utils import (AutoWeightsLoader, is_pp_missing_parameter,
48
make_empty_intermediate_tensors_factory, make_layers,
49
maybe_prefix)
@@ -313,7 +313,7 @@ class BloomModel(nn.Module):
313
return loaded_params
314
315
316
-class BloomForCausalLM(nn.Module, SupportsPP, SupportsV0Only, SupportsQuant):
316
+class BloomForCausalLM(nn.Module, SupportsPP, SupportsQuant):
317
318
def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
319
super().__init__()