Skip to content

Commit f943110

Browse files
authored
Merge branch 'develop' into doc
2 parents 0d39e23 + f4ce039 commit f943110

File tree

2 files changed

+2
-2
lines changed

2 files changed

+2
-2
lines changed

docs/supported_models.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -13,7 +13,7 @@ export FD_MODEL_CACHE=/ssd1/download_models
1313

1414
| Model Name | Context Length | Quantization | Minimum Deployment Resources | Notes |
1515
| :--------- | :------------- | :----------- | :-------------------------- | :---- |
16-
| baidu/ERNIE-4.5-VL-424B-A47B-Paddle | 32K/128K | WINT2 | 1*96G GPU VRAM/1T RAM | Chunked Prefill required for 128K |
16+
| baidu/ERNIE-4.5-VL-424B-A47B-Paddle | 32K/128K | WINT2 | 1*141G GPU VRAM/1T RAM | Chunked Prefill required for 128K |
1717
| baidu/ERNIE-4.5-VL-424B-A47B-Paddle | 32K/128K | WINT4 | 4*80G GPU VRAM/1T RAM | Chunked Prefill required for 128K |
1818
| baidu/ERNIE-4.5-VL-424B-A47B-Paddle | 32K/128K | WINT8 | 8*80G GPU VRAM/1T RAM | Chunked Prefill required for 128K |
1919
| baidu/ERNIE-4.5-300B-A47B-Paddle | 32K/128K | WINT4 | 4*64G GPU VRAM/600G RAM | Chunked Prefill required for 128K |

docs/zh/supported_models.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,7 @@ export FD_MODEL_CACHE=/ssd1/download_models
1414

1515
| 模型名 | 上下文长度 | 量化方式 | 最小部署资源 | 说明 |
1616
| :----- | :-------------- | :----------- |:----------- |:----------- |
17-
| baidu/ERNIE-4.5-VL-424B-A47B-Paddle | 32K/128K | WINT2 | 1卡*96G显存/1T内存 | 128K需要开启Chunked Prefill |
17+
| baidu/ERNIE-4.5-VL-424B-A47B-Paddle | 32K/128K | WINT2 | 1卡*141G显存/1T内存 | 128K需要开启Chunked Prefill |
1818
| baidu/ERNIE-4.5-VL-424B-A47B-Paddle | 32K/128K | WINT4 | 4卡*80G显存/1T内存 | 128K需要开启Chunked Prefill |
1919
| baidu/ERNIE-4.5-VL-424B-A47B-Paddle | 32K/128K | WINT8 | 8卡*80G显存/1T内存 | 128K需要开启Chunked Prefill |
2020
| baidu/ERNIE-4.5-300B-A47B-Paddle | 32K/128K | WINT4 | 4卡*64G显存/600G内存 | 128K需要开启Chunked Prefill |

0 commit comments

Comments
 (0)