ros/LLaMA-Factory
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update readme

Browse Source 
Former-commit-id: 1deda4750e0df6c46aeb33cf3f8b35baa537cc1d
This commit is contained in:
hiyouga
2024-12-23 14:08:59 +00:00
parent df42e438c1
commit a5346041bb
4 changed files with 85 additions and 39 deletions
Download Patch File Download Diff File Expand all files Collapse all files

106
src/llamafactory/extras/constants.py
View File

@@ -325,6 +325,7 @@ register_model_group(
models={
"Codestral-22B-v0.1-Chat": {
DownloadSource.DEFAULT: "mistralai/Codestral-22B-v0.1",
DownloadSource.MODELSCOPE: "swift/Codestral-22B-v0.1",
},
},
template="mistral",
@@ -420,15 +421,19 @@ register_model_group(
},
"DeepSeek-Coder-V2-16B-Base": {
DownloadSource.DEFAULT: "deepseek-ai/DeepSeek-Coder-V2-Lite-Base",
DownloadSource.MODELSCOPE: "deepseek-ai/DeepSeek-Coder-V2-Lite-Base",
},
"DeepSeek-Coder-V2-236B-Base": {
DownloadSource.DEFAULT: "deepseek-ai/DeepSeek-Coder-V2-Base",
DownloadSource.MODELSCOPE: "deepseek-ai/DeepSeek-Coder-V2-Base",
},
"DeepSeek-Coder-V2-16B-Instruct": {
DownloadSource.DEFAULT: "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
DownloadSource.MODELSCOPE: "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
},
"DeepSeek-Coder-V2-236B-Instruct": {
DownloadSource.DEFAULT: "deepseek-ai/DeepSeek-Coder-V2-Instruct",
DownloadSource.MODELSCOPE: "deepseek-ai/DeepSeek-Coder-V2-Instruct",
},
},
template="deepseek",
@@ -482,6 +487,7 @@ register_model_group(
},
"Falcon-11B": {
DownloadSource.DEFAULT: "tiiuae/falcon-11B",
DownloadSource.MODELSCOPE: "tiiuae/falcon-11B",
},
"Falcon-40B": {
DownloadSource.DEFAULT: "tiiuae/falcon-40b",
@@ -585,37 +591,69 @@ register_model_group(
register_model_group(
models={
"Granite-3.0-8B-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-8b-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-8b-instruct",
},
"Granite-3.0-8B-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-8b-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-8b-base",
},
"Granite-3.0-2B-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-2b-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-2b-instruct",
},
"Granite-3.0-2B-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-2b-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-2b-base",
},
"Granite-3.0-3B-A800M-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-3b-a800m-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-3b-a800m-instruct",
"Granite-3.0-1B-A400M-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-1b-a400m-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-1b-a400m-base",
},
"Granite-3.0-3B-A800M-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-3b-a800m-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-3b-a800m-base",
},
"Granite-3.0-2B-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-2b-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-2b-base",
},
"Granite-3.0-8B-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-8b-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-8b-base",
},
"Granite-3.0-1B-A400M-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-1b-a400m-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-1b-a400m-instruct",
},
"Granite-3.0-1B-A400M-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-1b-a400m-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-1b-a400m-base",
"Granite-3.0-3B-A800M-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-3b-a800m-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-3b-a800m-instruct",
},
"Granite-3.0-2B-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-2b-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-2b-instruct",
},
"Granite-3.0-8B-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.0-8b-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.0-8b-instruct",
},
"Granite-3.1-1B-A400M-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.1-1b-a400m-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.1-1b-a400m-base",
},
"Granite-3.1-3B-A800M-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.1-3b-a800m-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.1-3b-a800m-base",
},
"Granite-3.1-2B-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.1-2b-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.1-2b-base",
},
"Granite-3.1-8B-Base": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.1-8b-base",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.1-8b-base",
},
"Granite-3.1-1B-A400M-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.1-1b-a400m-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.1-1b-a400m-instruct",
},
"Granite-3.1-3B-A800M-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.1-3b-a800m-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.1-3b-a800m-instruct",
},
"Granite-3.1-2B-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.1-2b-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.1-2b-instruct",
},
"Granite-3.1-8B-Instruct": {
DownloadSource.DEFAULT: "ibm-granite/granite-3.1-8b-instruct",
DownloadSource.MODELSCOPE: "AI-ModelScope/granite-3.1-8b-instruct",
},
},
template="granite3",
@@ -624,14 +662,6 @@ register_model_group(
register_model_group(
models={
"Index-1.9B-Chat": {
DownloadSource.DEFAULT: "IndexTeam/Index-1.9B-Chat",
DownloadSource.MODELSCOPE: "IndexTeam/Index-1.9B-Chat",
},
"Index-1.9B-Character-Chat": {
DownloadSource.DEFAULT: "IndexTeam/Index-1.9B-Character",
DownloadSource.MODELSCOPE: "IndexTeam/Index-1.9B-Character",
},
"Index-1.9B-Base": {
DownloadSource.DEFAULT: "IndexTeam/Index-1.9B",
DownloadSource.MODELSCOPE: "IndexTeam/Index-1.9B",
@@ -640,6 +670,14 @@ register_model_group(
DownloadSource.DEFAULT: "IndexTeam/Index-1.9B-Pure",
DownloadSource.MODELSCOPE: "IndexTeam/Index-1.9B-Pure",
},
"Index-1.9B-Chat": {
DownloadSource.DEFAULT: "IndexTeam/Index-1.9B-Chat",
DownloadSource.MODELSCOPE: "IndexTeam/Index-1.9B-Chat",
},
"Index-1.9B-Character-Chat": {
DownloadSource.DEFAULT: "IndexTeam/Index-1.9B-Character",
DownloadSource.MODELSCOPE: "IndexTeam/Index-1.9B-Character",
},
"Index-1.9B-Chat-32K": {
DownloadSource.DEFAULT: "IndexTeam/Index-1.9B-32K",
DownloadSource.MODELSCOPE: "IndexTeam/Index-1.9B-32K",
@@ -1256,15 +1294,19 @@ register_model_group(
},
"PaliGemma2-3B-pt-448": {
DownloadSource.DEFAULT: "google/paligemma2-3b-pt-448",
DownloadSource.MODELSCOPE: "AI-ModelScope/paligemma2-3b-pt-448",
},
"PaliGemma2-3B-pt-896": {
DownloadSource.DEFAULT: "google/paligemma2-3b-pt-896",
DownloadSource.MODELSCOPE: "AI-ModelScope/paligemma2-3b-pt-896",
},
"PaliGemma2-10B-pt-224": {
DownloadSource.DEFAULT: "google/paligemma2-10b-pt-224",
DownloadSource.MODELSCOPE: "AI-ModelScope/paligemma2-10b-pt-224",
},
"PaliGemma2-10B-pt-448": {
DownloadSource.DEFAULT: "google/paligemma2-10b-pt-448",
DownloadSource.MODELSCOPE: "AI-ModelScope/paligemma2-10b-pt-448",
},
"PaliGemma2-10B-pt-896": {
DownloadSource.DEFAULT: "google/paligemma2-10b-pt-896",
@@ -1272,12 +1314,15 @@ register_model_group(
},
"PaliGemma2-28B-pt-224": {
DownloadSource.DEFAULT: "google/paligemma2-28b-pt-224",
DownloadSource.MODELSCOPE: "AI-ModelScope/paligemma2-28b-pt-224",
},
"PaliGemma2-28B-pt-448": {
DownloadSource.DEFAULT: "google/paligemma2-28b-pt-448",
DownloadSource.MODELSCOPE: "AI-ModelScope/paligemma2-28b-pt-448",
},
"PaliGemma2-28B-pt-896": {
DownloadSource.DEFAULT: "google/paligemma2-28b-pt-896",
DownloadSource.MODELSCOPE: "AI-ModelScope/paligemma2-28b-pt-896",
},
},
template="paligemma",
@@ -2006,6 +2051,7 @@ register_model_group(
models={
"Skywork-o1-Open-Llama-3.1-8B": {
DownloadSource.DEFAULT: "Skywork/Skywork-o1-Open-Llama-3.1-8B",
DownloadSource.MODELSCOPE: "AI-ModelScope/Skywork-o1-Open-Llama-3.1-8B",
}
},
template="skywork_o1",