Skip to content

Commit 3dfc3ab

Browse files
author
Judd
committed
add quantized InternLM models
1 parent adbfffa commit 3dfc3ab

File tree

1 file changed

+91
-0
lines changed

1 file changed

+91
-0
lines changed

scripts/models.json

Lines changed: 91 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,95 @@
11
{
2+
"internlm": {
3+
"brief": "InternLM2.5 is a 7B parameter model tailored for practical scenarios with outstanding reasoning capability.",
4+
"default": "7b",
5+
"license": "Apache License Version 2.0",
6+
"variants": {
7+
"7b": {
8+
"default": "q8",
9+
"quantized": {
10+
"q8": {
11+
"size": 8223436400,
12+
"url": "chatllm_quantized_internlm/internlm2.5-7b.bin"
13+
}
14+
}
15+
},
16+
"7b-1m": {
17+
"default": "q8",
18+
"quantized": {
19+
"q8": {
20+
"size": 8223436400,
21+
"url": "chatllm_quantized_internlm/internlm2.5-7b-1m.bin"
22+
}
23+
}
24+
}
25+
}
26+
},
27+
"internlm2": {
28+
"brief": "The second generation of the InternLM model.",
29+
"default": "1.8b",
30+
"license": "https://huggingface.co/internlm/internlm2-7b#open-source-license",
31+
"variants": {
32+
"1.8b": {
33+
"default": "q8",
34+
"quantized": {
35+
"q8": {
36+
"size": 2008808560,
37+
"url": "chatllm_quantized_internlm/internlm2-1.8B.bin"
38+
}
39+
}
40+
},
41+
"8b": {
42+
"default": "q8",
43+
"quantized": {
44+
"q8": {
45+
"size": 8223436400,
46+
"url": "chatllm_quantized_internlm/internlm2-chat-8b-new.bin"
47+
}
48+
}
49+
},
50+
"20b": {
51+
"default": "q8",
52+
"quantized": {
53+
"q8": {
54+
"size": 21105570416,
55+
"url": "chatllm_quantized_internlm/internlm2-20b.bin"
56+
}
57+
}
58+
}
59+
}
60+
},
61+
"internlm2-math": {
62+
"brief": "State-of-the-art bilingual open-sourced Math reasoning LLMs. A solver, prover, verifier, augmenter.",
63+
"default": "1.8b",
64+
"license": "Unknown. See https://huggingface.co/internlm/internlm2-math-plus-1_8b",
65+
"variants": {
66+
"1.8b": {
67+
"default": "q8",
68+
"quantized": {
69+
"q8": {
70+
"size": 2008808560,
71+
"url": "chatllm_quantized_internlm/internlm2-math-plus-1_8b.bin"
72+
}
73+
}
74+
}
75+
}
76+
},
77+
"internlm1": {
78+
"brief": "InternLM has open-sourced a 7 billion parameter base model and a chat model tailored for practical scenarios.",
79+
"default": "1.8b",
80+
"license": "https://huggingface.co/internlm/internlm-chat-7b#open-source-license",
81+
"variants": {
82+
"1.8b": {
83+
"default": "q8",
84+
"quantized": {
85+
"q8": {
86+
"size": 7783400944,
87+
"url": "chatllm_quantized_internlm/internlmv1.1_7b.bin"
88+
}
89+
}
90+
}
91+
}
92+
},
293
"llm-compiler": {
394
"brief": "LLM Compiler is a state-of-the-art LLM that builds upon Code Llama with improved performance for code optimization and compiler reasoning.",
495
"default": "7b",

0 commit comments

Comments
 (0)