add deepseek3 model

Former-commit-id: 611779d412f31e25b1ed38049050eee2da61dde5
This commit is contained in:
hiyouga
2024-12-30 13:38:30 +00:00
parent e3f95abca7
commit b2e4f11602
4 changed files with 32 additions and 2 deletions

View File

@@ -619,9 +619,8 @@ _register_template(
_register_template(
name="deepseek_v2.5",
name="deepseek3",
format_user=StringFormatter(slots=["<User>{{content}}<Assistant>"]),
format_system=StringFormatter(slots=["{{content}}"]),
format_prefix=EmptyFormatter(slots=[{"bos_token"}]),
)

View File

@@ -448,6 +448,7 @@ register_model_group(
},
"DeepSeek-Coder-7B-Base": {
DownloadSource.DEFAULT: "deepseek-ai/deepseek-coder-7b-base-v1.5",
DownloadSource.MODELSCOPE: "deepseek-ai/deepseek-coder-7b-base-v1.5",
},
"DeepSeek-Coder-33B-Base": {
DownloadSource.DEFAULT: "deepseek-ai/deepseek-coder-33b-base",
@@ -459,6 +460,7 @@ register_model_group(
},
"DeepSeek-Coder-7B-Instruct": {
DownloadSource.DEFAULT: "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
DownloadSource.MODELSCOPE: "deepseek-ai/deepseek-coder-7b-instruct-v1.5",
},
"DeepSeek-Coder-33B-Instruct": {
DownloadSource.DEFAULT: "deepseek-ai/deepseek-coder-33b-instruct",
@@ -469,6 +471,33 @@ register_model_group(
)
register_model_group(
models={
"DeepSeek-V2-236B-Chat-0628": {
DownloadSource.DEFAULT: "deepseek-ai/DeepSeek-V2-Chat-0628",
DownloadSource.MODELSCOPE: "deepseek-ai/DeepSeek-V2-Chat-0628",
},
"DeepSeek-V2.5-236B-Chat": {
DownloadSource.DEFAULT: "deepseek-ai/DeepSeek-V2.5",
DownloadSource.MODELSCOPE: "deepseek-ai/DeepSeek-V2.5",
},
"DeepSeek-V2.5-236B-Chat-1210": {
DownloadSource.DEFAULT: "deepseek-ai/DeepSeek-V2.5-1210",
DownloadSource.MODELSCOPE: "deepseek-ai/DeepSeek-V2.5-1210",
},
"DeepSeek-V3-685B-Base": {
DownloadSource.DEFAULT: "deepseek-ai/DeepSeek-V3-Base",
DownloadSource.MODELSCOPE: "deepseek-ai/DeepSeek-V3-Base",
},
"DeepSeek-V3-685B-Chat": {
DownloadSource.DEFAULT: "deepseek-ai/DeepSeek-V3",
DownloadSource.MODELSCOPE: "deepseek-ai/DeepSeek-V3",
},
},
template="deepseek3",
)
register_model_group(
models={
"EXAONE-3.0-7.8B-Instruct": {