Upload 2 files
Browse files- LICENSE.txt +201 -0
- README.md +197 -22
LICENSE.txt
ADDED
@@ -0,0 +1,201 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Apache License
|
2 |
+
Version 2.0, January 2004
|
3 |
+
http://www.apache.org/licenses/
|
4 |
+
|
5 |
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
6 |
+
|
7 |
+
1. Definitions.
|
8 |
+
|
9 |
+
"License" shall mean the terms and conditions for use, reproduction,
|
10 |
+
and distribution as defined by Sections 1 through 9 of this document.
|
11 |
+
|
12 |
+
"Licensor" shall mean the copyright owner or entity authorized by
|
13 |
+
the copyright owner that is granting the License.
|
14 |
+
|
15 |
+
"Legal Entity" shall mean the union of the acting entity and all
|
16 |
+
other entities that control, are controlled by, or are under common
|
17 |
+
control with that entity. For the purposes of this definition,
|
18 |
+
"control" means (i) the power, direct or indirect, to cause the
|
19 |
+
direction or management of such entity, whether by contract or
|
20 |
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
21 |
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
22 |
+
|
23 |
+
"You" (or "Your") shall mean an individual or Legal Entity
|
24 |
+
exercising permissions granted by this License.
|
25 |
+
|
26 |
+
"Source" form shall mean the preferred form for making modifications,
|
27 |
+
including but not limited to software source code, documentation
|
28 |
+
source, and configuration files.
|
29 |
+
|
30 |
+
"Object" form shall mean any form resulting from mechanical
|
31 |
+
transformation or translation of a Source form, including but
|
32 |
+
not limited to compiled object code, generated documentation,
|
33 |
+
and conversions to other media types.
|
34 |
+
|
35 |
+
"Work" shall mean the work of authorship, whether in Source or
|
36 |
+
Object form, made available under the License, as indicated by a
|
37 |
+
copyright notice that is included in or attached to the work
|
38 |
+
(an example is provided in the Appendix below).
|
39 |
+
|
40 |
+
"Derivative Works" shall mean any work, whether in Source or Object
|
41 |
+
form, that is based on (or derived from) the Work and for which the
|
42 |
+
editorial revisions, annotations, elaborations, or other modifications
|
43 |
+
represent, as a whole, an original work of authorship. For the purposes
|
44 |
+
of this License, Derivative Works shall not include works that remain
|
45 |
+
separable from, or merely link (or bind by name) to the interfaces of,
|
46 |
+
the Work and Derivative Works thereof.
|
47 |
+
|
48 |
+
"Contribution" shall mean any work of authorship, including
|
49 |
+
the original version of the Work and any modifications or additions
|
50 |
+
to that Work or Derivative Works thereof, that is intentionally
|
51 |
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
52 |
+
or by an individual or Legal Entity authorized to submit on behalf of
|
53 |
+
the copyright owner. For the purposes of this definition, "submitted"
|
54 |
+
means any form of electronic, verbal, or written communication sent
|
55 |
+
to the Licensor or its representatives, including but not limited to
|
56 |
+
communication on electronic mailing lists, source code control systems,
|
57 |
+
and issue tracking systems that are managed by, or on behalf of, the
|
58 |
+
Licensor for the purpose of discussing and improving the Work, but
|
59 |
+
excluding communication that is conspicuously marked or otherwise
|
60 |
+
designated in writing by the copyright owner as "Not a Contribution."
|
61 |
+
|
62 |
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
63 |
+
on behalf of whom a Contribution has been received by Licensor and
|
64 |
+
subsequently incorporated within the Work.
|
65 |
+
|
66 |
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
67 |
+
this License, each Contributor hereby grants to You a perpetual,
|
68 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
69 |
+
copyright license to reproduce, prepare Derivative Works of,
|
70 |
+
publicly display, publicly perform, sublicense, and distribute the
|
71 |
+
Work and such Derivative Works in Source or Object form.
|
72 |
+
|
73 |
+
3. Grant of Patent License. Subject to the terms and conditions of
|
74 |
+
this License, each Contributor hereby grants to You a perpetual,
|
75 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
76 |
+
(except as stated in this section) patent license to make, have made,
|
77 |
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
78 |
+
where such license applies only to those patent claims licensable
|
79 |
+
by such Contributor that are necessarily infringed by their
|
80 |
+
Contribution(s) alone or by combination of their Contribution(s)
|
81 |
+
with the Work to which such Contribution(s) was submitted. If You
|
82 |
+
institute patent litigation against any entity (including a
|
83 |
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
84 |
+
or a Contribution incorporated within the Work constitutes direct
|
85 |
+
or contributory patent infringement, then any patent licenses
|
86 |
+
granted to You under this License for that Work shall terminate
|
87 |
+
as of the date such litigation is filed.
|
88 |
+
|
89 |
+
4. Redistribution. You may reproduce and distribute copies of the
|
90 |
+
Work or Derivative Works thereof in any medium, with or without
|
91 |
+
modifications, and in Source or Object form, provided that You
|
92 |
+
meet the following conditions:
|
93 |
+
|
94 |
+
(a) You must give any other recipients of the Work or
|
95 |
+
Derivative Works a copy of this License; and
|
96 |
+
|
97 |
+
(b) You must cause any modified files to carry prominent notices
|
98 |
+
stating that You changed the files; and
|
99 |
+
|
100 |
+
(c) You must retain, in the Source form of any Derivative Works
|
101 |
+
that You distribute, all copyright, patent, trademark, and
|
102 |
+
attribution notices from the Source form of the Work,
|
103 |
+
excluding those notices that do not pertain to any part of
|
104 |
+
the Derivative Works; and
|
105 |
+
|
106 |
+
(d) If the Work includes a "NOTICE" text file as part of its
|
107 |
+
distribution, then any Derivative Works that You distribute must
|
108 |
+
include a readable copy of the attribution notices contained
|
109 |
+
within such NOTICE file, excluding those notices that do not
|
110 |
+
pertain to any part of the Derivative Works, in at least one
|
111 |
+
of the following places: within a NOTICE text file distributed
|
112 |
+
as part of the Derivative Works; within the Source form or
|
113 |
+
documentation, if provided along with the Derivative Works; or,
|
114 |
+
within a display generated by the Derivative Works, if and
|
115 |
+
wherever such third-party notices normally appear. The contents
|
116 |
+
of the NOTICE file are for informational purposes only and
|
117 |
+
do not modify the License. You may add Your own attribution
|
118 |
+
notices within Derivative Works that You distribute, alongside
|
119 |
+
or as an addendum to the NOTICE text from the Work, provided
|
120 |
+
that such additional attribution notices cannot be construed
|
121 |
+
as modifying the License.
|
122 |
+
|
123 |
+
You may add Your own copyright statement to Your modifications and
|
124 |
+
may provide additional or different license terms and conditions
|
125 |
+
for use, reproduction, or distribution of Your modifications, or
|
126 |
+
for any such Derivative Works as a whole, provided Your use,
|
127 |
+
reproduction, and distribution of the Work otherwise complies with
|
128 |
+
the conditions stated in this License.
|
129 |
+
|
130 |
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
131 |
+
any Contribution intentionally submitted for inclusion in the Work
|
132 |
+
by You to the Licensor shall be under the terms and conditions of
|
133 |
+
this License, without any additional terms or conditions.
|
134 |
+
Notwithstanding the above, nothing herein shall supersede or modify
|
135 |
+
the terms of any separate license agreement you may have executed
|
136 |
+
with Licensor regarding such Contributions.
|
137 |
+
|
138 |
+
6. Trademarks. This License does not grant permission to use the trade
|
139 |
+
names, trademarks, service marks, or product names of the Licensor,
|
140 |
+
except as required for reasonable and customary use in describing the
|
141 |
+
origin of the Work and reproducing the content of the NOTICE file.
|
142 |
+
|
143 |
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
144 |
+
agreed to in writing, Licensor provides the Work (and each
|
145 |
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
146 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
147 |
+
implied, including, without limitation, any warranties or conditions
|
148 |
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
149 |
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
150 |
+
appropriateness of using or redistributing the Work and assume any
|
151 |
+
risks associated with Your exercise of permissions under this License.
|
152 |
+
|
153 |
+
8. Limitation of Liability. In no event and under no legal theory,
|
154 |
+
whether in tort (including negligence), contract, or otherwise,
|
155 |
+
unless required by applicable law (such as deliberate and grossly
|
156 |
+
negligent acts) or agreed to in writing, shall any Contributor be
|
157 |
+
liable to You for damages, including any direct, indirect, special,
|
158 |
+
incidental, or consequential damages of any character arising as a
|
159 |
+
result of this License or out of the use or inability to use the
|
160 |
+
Work (including but not limited to damages for loss of goodwill,
|
161 |
+
work stoppage, computer failure or malfunction, or any and all
|
162 |
+
other commercial damages or losses), even if such Contributor
|
163 |
+
has been advised of the possibility of such damages.
|
164 |
+
|
165 |
+
9. Accepting Warranty or Additional Liability. While redistributing
|
166 |
+
the Work or Derivative Works thereof, You may choose to offer,
|
167 |
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
168 |
+
or other liability obligations and/or rights consistent with this
|
169 |
+
License. However, in accepting such obligations, You may act only
|
170 |
+
on Your own behalf and on Your sole responsibility, not on behalf
|
171 |
+
of any other Contributor, and only if You agree to indemnify,
|
172 |
+
defend, and hold each Contributor harmless for any liability
|
173 |
+
incurred by, or claims asserted against, such Contributor by reason
|
174 |
+
of your accepting any such warranty or additional liability.
|
175 |
+
|
176 |
+
END OF TERMS AND CONDITIONS
|
177 |
+
|
178 |
+
APPENDIX: How to apply the Apache License to your work.
|
179 |
+
|
180 |
+
To apply the Apache License to your work, attach the following
|
181 |
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
182 |
+
replaced with your own identifying information. (Don't include
|
183 |
+
the brackets!) The text should be enclosed in the appropriate
|
184 |
+
comment syntax for the file format. We also recommend that a
|
185 |
+
file or class name and description of purpose be included on the
|
186 |
+
same "printed page" as the copyright notice for easier
|
187 |
+
identification within third-party archives.
|
188 |
+
|
189 |
+
Copyright 2023-2024 Shanghai AI Laboratory
|
190 |
+
|
191 |
+
Licensed under the Apache License, Version 2.0 (the "License");
|
192 |
+
you may not use this file except in compliance with the License.
|
193 |
+
You may obtain a copy of the License at
|
194 |
+
|
195 |
+
http://www.apache.org/licenses/LICENSE-2.0
|
196 |
+
|
197 |
+
Unless required by applicable law or agreed to in writing, software
|
198 |
+
distributed under the License is distributed on an "AS IS" BASIS,
|
199 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
200 |
+
See the License for the specific language governing permissions and
|
201 |
+
limitations under the License.
|
README.md
CHANGED
@@ -1,5 +1,6 @@
|
|
1 |
---
|
2 |
-
|
|
|
3 |
---
|
4 |
# InternLM
|
5 |
|
@@ -89,7 +90,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
|
|
89 |
model_dir = "internlm/internlm3-8b-instruct"
|
90 |
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
|
91 |
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
|
92 |
-
model = AutoModelForCausalLM.from_pretrained(model_dir, trust_remote_code=True, torch_dtype=torch.
|
93 |
# (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
|
94 |
# InternLM3 8B in 4bit will cost nearly 8GB GPU memory.
|
95 |
# pip install -U bitsandbytes
|
@@ -104,7 +105,7 @@ messages = [
|
|
104 |
{"role": "system", "content": system_prompt},
|
105 |
{"role": "user", "content": "Please tell me five scenic spots in Shanghai"},
|
106 |
]
|
107 |
-
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
|
108 |
|
109 |
generated_ids = model.generate(tokenized_chat, max_new_tokens=1024, temperature=1, repetition_penalty=1.005, top_k=40, top_p=0.8)
|
110 |
|
@@ -113,7 +114,7 @@ generated_ids = [
|
|
113 |
]
|
114 |
prompt = tokenizer.batch_decode(tokenized_chat)[0]
|
115 |
print(prompt)
|
116 |
-
response = tokenizer.batch_decode(generated_ids)[0]
|
117 |
print(response)
|
118 |
```
|
119 |
|
@@ -160,7 +161,47 @@ Find more details in the [LMDeploy documentation](https://lmdeploy.readthedocs.i
|
|
160 |
|
161 |
#### Ollama inference
|
162 |
|
163 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
164 |
|
165 |
#### vLLM inference
|
166 |
|
@@ -168,7 +209,11 @@ We are still working on merging the PR(https://github.com/vllm-project/vllm/pull
|
|
168 |
|
169 |
```python
|
170 |
git clone -b support-internlm3 https://github.com/RunningLeon/vllm.git
|
171 |
-
|
|
|
|
|
|
|
|
|
172 |
```
|
173 |
|
174 |
inference code:
|
@@ -270,7 +315,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
|
|
270 |
model_dir = "internlm/internlm3-8b-instruct"
|
271 |
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
|
272 |
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
|
273 |
-
model = AutoModelForCausalLM.from_pretrained(model_dir, trust_remote_code=True, torch_dtype=torch.
|
274 |
# (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
|
275 |
# InternLM3 8B in 4bit will cost nearly 8GB GPU memory.
|
276 |
# pip install -U bitsandbytes
|
@@ -282,7 +327,7 @@ messages = [
|
|
282 |
{"role": "system", "content": thinking_system_prompt},
|
283 |
{"role": "user", "content": "Given the function\(f(x)=\mathrm{e}^{x}-ax - a^{3}\),\n(1) When \(a = 1\), find the equation of the tangent line to the curve \(y = f(x)\) at the point \((1,f(1))\).\n(2) If \(f(x)\) has a local minimum and the minimum value is less than \(0\), determine the range of values for \(a\)."},
|
284 |
]
|
285 |
-
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
|
286 |
|
287 |
generated_ids = model.generate(tokenized_chat, max_new_tokens=8192)
|
288 |
|
@@ -291,7 +336,7 @@ generated_ids = [
|
|
291 |
]
|
292 |
prompt = tokenizer.batch_decode(tokenized_chat)[0]
|
293 |
print(prompt)
|
294 |
-
response = tokenizer.batch_decode(generated_ids)[0]
|
295 |
print(response)
|
296 |
```
|
297 |
#### LMDeploy inference
|
@@ -321,14 +366,56 @@ print(response)
|
|
321 |
|
322 |
#### Ollama inference
|
323 |
|
324 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
325 |
|
326 |
#### vLLM inference
|
327 |
|
328 |
We are still working on merging the PR(https://github.com/vllm-project/vllm/pull/12037) into vLLM. In the meantime, please use the following PR link to install it manually.
|
329 |
```python
|
330 |
git clone https://github.com/RunningLeon/vllm.git
|
331 |
-
|
|
|
|
|
|
|
|
|
332 |
```
|
333 |
|
334 |
inference code
|
@@ -438,7 +525,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
|
|
438 |
model_dir = "internlm/internlm3-8b-instruct"
|
439 |
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
|
440 |
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
|
441 |
-
model = AutoModelForCausalLM.from_pretrained(model_dir, trust_remote_code=True, torch_dtype=torch.
|
442 |
# (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
|
443 |
# InternLM3 8B in 4bit will cost nearly 8GB GPU memory.
|
444 |
# pip install -U bitsandbytes
|
@@ -453,7 +540,7 @@ messages = [
|
|
453 |
{"role": "system", "content": system_prompt},
|
454 |
{"role": "user", "content": "Please tell me five scenic spots in Shanghai"},
|
455 |
]
|
456 |
-
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
|
457 |
|
458 |
generated_ids = model.generate(tokenized_chat, max_new_tokens=1024, temperature=1, repetition_penalty=1.005, top_k=40, top_p=0.8)
|
459 |
|
@@ -462,7 +549,7 @@ generated_ids = [
|
|
462 |
]
|
463 |
prompt = tokenizer.batch_decode(tokenized_chat)[0]
|
464 |
print(prompt)
|
465 |
-
response = tokenizer.batch_decode(generated_ids)[0]
|
466 |
print(response)
|
467 |
```
|
468 |
|
@@ -510,7 +597,49 @@ curl http://localhost:23333/v1/chat/completions \
|
|
510 |
|
511 |
##### Ollama 推理
|
512 |
|
513 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
514 |
|
515 |
##### vLLM 推理
|
516 |
|
@@ -518,7 +647,11 @@ TODO
|
|
518 |
|
519 |
```python
|
520 |
git clone https://github.com/RunningLeon/vllm.git
|
521 |
-
|
|
|
|
|
|
|
|
|
522 |
```
|
523 |
|
524 |
推理代码
|
@@ -619,7 +752,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
|
|
619 |
model_dir = "internlm/internlm3-8b-instruct"
|
620 |
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
|
621 |
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
|
622 |
-
model = AutoModelForCausalLM.from_pretrained(model_dir, trust_remote_code=True, torch_dtype=torch.
|
623 |
# (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
|
624 |
# InternLM3 8B in 4bit will cost nearly 8GB GPU memory.
|
625 |
# pip install -U bitsandbytes
|
@@ -631,7 +764,7 @@ messages = [
|
|
631 |
{"role": "system", "content": thinking_system_prompt},
|
632 |
{"role": "user", "content": "已知函数\(f(x)=\mathrm{e}^{x}-ax - a^{3}\)。\n(1)当\(a = 1\)时,求曲线\(y = f(x)\)在点\((1,f(1))\)处的切线方程;\n(2)若\(f(x)\)有极小值,且极小值小于\(0\),求\(a\)的取值范围。"},
|
633 |
]
|
634 |
-
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
|
635 |
|
636 |
generated_ids = model.generate(tokenized_chat, max_new_tokens=8192)
|
637 |
|
@@ -640,7 +773,7 @@ generated_ids = [
|
|
640 |
]
|
641 |
prompt = tokenizer.batch_decode(tokenized_chat)[0]
|
642 |
print(prompt)
|
643 |
-
response = tokenizer.batch_decode(generated_ids)[0]
|
644 |
print(response)
|
645 |
```
|
646 |
##### LMDeploy 推理
|
@@ -670,7 +803,45 @@ print(response)
|
|
670 |
|
671 |
##### Ollama 推理
|
672 |
|
673 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
674 |
|
675 |
##### vLLM 推理
|
676 |
|
@@ -678,7 +849,11 @@ TODO
|
|
678 |
|
679 |
```python
|
680 |
git clone https://github.com/RunningLeon/vllm.git
|
681 |
-
|
|
|
|
|
|
|
|
|
682 |
```
|
683 |
|
684 |
推理代码
|
@@ -728,4 +903,4 @@ print(outputs)
|
|
728 |
archivePrefix={arXiv},
|
729 |
primaryClass={cs.CL}
|
730 |
}
|
731 |
-
```
|
|
|
1 |
---
|
2 |
+
pipeline_tag: text-generation
|
3 |
+
license: other
|
4 |
---
|
5 |
# InternLM
|
6 |
|
|
|
90 |
model_dir = "internlm/internlm3-8b-instruct"
|
91 |
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
|
92 |
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
|
93 |
+
model = AutoModelForCausalLM.from_pretrained(model_dir, trust_remote_code=True, torch_dtype=torch.bfloat16).cuda()
|
94 |
# (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
|
95 |
# InternLM3 8B in 4bit will cost nearly 8GB GPU memory.
|
96 |
# pip install -U bitsandbytes
|
|
|
105 |
{"role": "system", "content": system_prompt},
|
106 |
{"role": "user", "content": "Please tell me five scenic spots in Shanghai"},
|
107 |
]
|
108 |
+
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda")
|
109 |
|
110 |
generated_ids = model.generate(tokenized_chat, max_new_tokens=1024, temperature=1, repetition_penalty=1.005, top_k=40, top_p=0.8)
|
111 |
|
|
|
114 |
]
|
115 |
prompt = tokenizer.batch_decode(tokenized_chat)[0]
|
116 |
print(prompt)
|
117 |
+
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
|
118 |
print(response)
|
119 |
```
|
120 |
|
|
|
161 |
|
162 |
#### Ollama inference
|
163 |
|
164 |
+
First install ollama,
|
165 |
+
|
166 |
+
```python
|
167 |
+
# install ollama
|
168 |
+
curl -fsSL https://ollama.com/install.sh | sh
|
169 |
+
# fetch model
|
170 |
+
ollama pull internlm/internlm3-8b-instruct
|
171 |
+
# install
|
172 |
+
pip install ollama
|
173 |
+
```
|
174 |
+
|
175 |
+
inference code,
|
176 |
+
|
177 |
+
```python
|
178 |
+
import ollama
|
179 |
+
|
180 |
+
system_prompt = """You are an AI assistant whose name is InternLM (书生·浦语).
|
181 |
+
- InternLM (书生·浦语) is a conversational language model that is developed by Shanghai AI Laboratory (上海人工智能实验室). It is designed to be helpful, honest, and harmless.
|
182 |
+
- InternLM (书生·浦语) can understand and communicate fluently in the language chosen by the user such as English and 中文."""
|
183 |
+
|
184 |
+
messages = [
|
185 |
+
{
|
186 |
+
"role": "system",
|
187 |
+
"content": system_prompt,
|
188 |
+
},
|
189 |
+
{
|
190 |
+
"role": "user",
|
191 |
+
"content": "Please tell me five scenic spots in Shanghai"
|
192 |
+
},
|
193 |
+
]
|
194 |
+
|
195 |
+
stream = ollama.chat(
|
196 |
+
model='internlm/internlm3-8b-instruct',
|
197 |
+
messages=messages,
|
198 |
+
stream=True,
|
199 |
+
)
|
200 |
+
|
201 |
+
for chunk in stream:
|
202 |
+
print(chunk['message']['content'], end='', flush=True)
|
203 |
+
```
|
204 |
+
|
205 |
|
206 |
#### vLLM inference
|
207 |
|
|
|
209 |
|
210 |
```python
|
211 |
git clone -b support-internlm3 https://github.com/RunningLeon/vllm.git
|
212 |
+
# and then follow https://docs.vllm.ai/en/latest/getting_started/installation/gpu/index.html#build-wheel-from-source to install
|
213 |
+
cd vllm
|
214 |
+
python use_existing_torch.py
|
215 |
+
pip install -r requirements-build.txt
|
216 |
+
pip install -e . --no-build-isolatio
|
217 |
```
|
218 |
|
219 |
inference code:
|
|
|
315 |
model_dir = "internlm/internlm3-8b-instruct"
|
316 |
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
|
317 |
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
|
318 |
+
model = AutoModelForCausalLM.from_pretrained(model_dir, trust_remote_code=True, torch_dtype=torch.bfloat16).cuda()
|
319 |
# (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
|
320 |
# InternLM3 8B in 4bit will cost nearly 8GB GPU memory.
|
321 |
# pip install -U bitsandbytes
|
|
|
327 |
{"role": "system", "content": thinking_system_prompt},
|
328 |
{"role": "user", "content": "Given the function\(f(x)=\mathrm{e}^{x}-ax - a^{3}\),\n(1) When \(a = 1\), find the equation of the tangent line to the curve \(y = f(x)\) at the point \((1,f(1))\).\n(2) If \(f(x)\) has a local minimum and the minimum value is less than \(0\), determine the range of values for \(a\)."},
|
329 |
]
|
330 |
+
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda")
|
331 |
|
332 |
generated_ids = model.generate(tokenized_chat, max_new_tokens=8192)
|
333 |
|
|
|
336 |
]
|
337 |
prompt = tokenizer.batch_decode(tokenized_chat)[0]
|
338 |
print(prompt)
|
339 |
+
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
|
340 |
print(response)
|
341 |
```
|
342 |
#### LMDeploy inference
|
|
|
366 |
|
367 |
#### Ollama inference
|
368 |
|
369 |
+
First install ollama,
|
370 |
+
|
371 |
+
```python
|
372 |
+
# install ollama
|
373 |
+
curl -fsSL https://ollama.com/install.sh | sh
|
374 |
+
# fetch model
|
375 |
+
ollama pull internlm/internlm3-8b-instruct
|
376 |
+
# install
|
377 |
+
pip install ollama
|
378 |
+
```
|
379 |
+
|
380 |
+
inference code,
|
381 |
+
|
382 |
+
```python
|
383 |
+
import ollama
|
384 |
+
|
385 |
+
messages = [
|
386 |
+
{
|
387 |
+
"role": "system",
|
388 |
+
"content": thinking_system_prompt,
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"role": "user",
|
392 |
+
"content": "Given the function\(f(x)=\mathrm{e}^{x}-ax - a^{3}\),\n(1) When \(a = 1\), find the equation of the tangent line to the curve \(y = f(x)\) at the point \((1,f(1))\).\n(2) If \(f(x)\) has a local minimum and the minimum value is less than \(0\), determine the range of values for \(a\)."
|
393 |
+
},
|
394 |
+
]
|
395 |
+
|
396 |
+
stream = ollama.chat(
|
397 |
+
model='internlm/internlm3-8b-instruct',
|
398 |
+
messages=messages,
|
399 |
+
stream=True,
|
400 |
+
)
|
401 |
+
|
402 |
+
for chunk in stream:
|
403 |
+
print(chunk['message']['content'], end='', flush=True)
|
404 |
+
```
|
405 |
+
|
406 |
+
|
407 |
+
####
|
408 |
|
409 |
#### vLLM inference
|
410 |
|
411 |
We are still working on merging the PR(https://github.com/vllm-project/vllm/pull/12037) into vLLM. In the meantime, please use the following PR link to install it manually.
|
412 |
```python
|
413 |
git clone https://github.com/RunningLeon/vllm.git
|
414 |
+
# and then follow https://docs.vllm.ai/en/latest/getting_started/installation/gpu/index.html#build-wheel-from-source to install
|
415 |
+
cd vllm
|
416 |
+
python use_existing_torch.py
|
417 |
+
pip install -r requirements-build.txt
|
418 |
+
pip install -e . --no-build-isolatio
|
419 |
```
|
420 |
|
421 |
inference code
|
|
|
525 |
model_dir = "internlm/internlm3-8b-instruct"
|
526 |
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
|
527 |
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
|
528 |
+
model = AutoModelForCausalLM.from_pretrained(model_dir, trust_remote_code=True, torch_dtype=torch.bfloat16).cuda()
|
529 |
# (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
|
530 |
# InternLM3 8B in 4bit will cost nearly 8GB GPU memory.
|
531 |
# pip install -U bitsandbytes
|
|
|
540 |
{"role": "system", "content": system_prompt},
|
541 |
{"role": "user", "content": "Please tell me five scenic spots in Shanghai"},
|
542 |
]
|
543 |
+
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda")
|
544 |
|
545 |
generated_ids = model.generate(tokenized_chat, max_new_tokens=1024, temperature=1, repetition_penalty=1.005, top_k=40, top_p=0.8)
|
546 |
|
|
|
549 |
]
|
550 |
prompt = tokenizer.batch_decode(tokenized_chat)[0]
|
551 |
print(prompt)
|
552 |
+
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
|
553 |
print(response)
|
554 |
```
|
555 |
|
|
|
597 |
|
598 |
##### Ollama 推理
|
599 |
|
600 |
+
准备工作
|
601 |
+
|
602 |
+
```python
|
603 |
+
# install ollama
|
604 |
+
curl -fsSL https://ollama.com/install.sh | sh
|
605 |
+
# fetch 模型
|
606 |
+
ollama pull internlm/internlm3-8b-instruct
|
607 |
+
# install python库
|
608 |
+
pip install ollama
|
609 |
+
```
|
610 |
+
|
611 |
+
推理代码
|
612 |
+
|
613 |
+
```python
|
614 |
+
import ollama
|
615 |
+
|
616 |
+
system_prompt = """You are an AI assistant whose name is InternLM (书生·浦语).
|
617 |
+
- InternLM (书生·浦语) is a conversational language model that is developed by Shanghai AI Laboratory (上海人工智能实验室). It is designed to be helpful, honest, and harmless.
|
618 |
+
- InternLM (书生·浦语) can understand and communicate fluently in the language chosen by the user such as English and 中文."""
|
619 |
+
|
620 |
+
messages = [
|
621 |
+
{
|
622 |
+
"role": "system",
|
623 |
+
"content": system_prompt,
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"role": "user",
|
627 |
+
"content": "Please tell me five scenic spots in Shanghai"
|
628 |
+
},
|
629 |
+
]
|
630 |
+
|
631 |
+
stream = ollama.chat(
|
632 |
+
model='internlm/internlm3-8b-instruct',
|
633 |
+
messages=messages,
|
634 |
+
stream=True,
|
635 |
+
)
|
636 |
+
|
637 |
+
for chunk in stream:
|
638 |
+
print(chunk['message']['content'], end='', flush=True)
|
639 |
+
```
|
640 |
+
|
641 |
+
|
642 |
+
####
|
643 |
|
644 |
##### vLLM 推理
|
645 |
|
|
|
647 |
|
648 |
```python
|
649 |
git clone https://github.com/RunningLeon/vllm.git
|
650 |
+
# and then follow https://docs.vllm.ai/en/latest/getting_started/installation/gpu/index.html#build-wheel-from-source to install
|
651 |
+
cd vllm
|
652 |
+
python use_existing_torch.py
|
653 |
+
pip install -r requirements-build.txt
|
654 |
+
pip install -e . --no-build-isolatio
|
655 |
```
|
656 |
|
657 |
推理代码
|
|
|
752 |
model_dir = "internlm/internlm3-8b-instruct"
|
753 |
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
|
754 |
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
|
755 |
+
model = AutoModelForCausalLM.from_pretrained(model_dir, trust_remote_code=True, torch_dtype=torch.bfloat16).cuda()
|
756 |
# (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
|
757 |
# InternLM3 8B in 4bit will cost nearly 8GB GPU memory.
|
758 |
# pip install -U bitsandbytes
|
|
|
764 |
{"role": "system", "content": thinking_system_prompt},
|
765 |
{"role": "user", "content": "已知函数\(f(x)=\mathrm{e}^{x}-ax - a^{3}\)。\n(1)当\(a = 1\)时,求曲线\(y = f(x)\)在点\((1,f(1))\)处的切线方程;\n(2)若\(f(x)\)有极小值,且极小值小于\(0\),求\(a\)的取值范围。"},
|
766 |
]
|
767 |
+
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda")
|
768 |
|
769 |
generated_ids = model.generate(tokenized_chat, max_new_tokens=8192)
|
770 |
|
|
|
773 |
]
|
774 |
prompt = tokenizer.batch_decode(tokenized_chat)[0]
|
775 |
print(prompt)
|
776 |
+
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
|
777 |
print(response)
|
778 |
```
|
779 |
##### LMDeploy 推理
|
|
|
803 |
|
804 |
##### Ollama 推理
|
805 |
|
806 |
+
准备工作
|
807 |
+
|
808 |
+
```python
|
809 |
+
# install ollama
|
810 |
+
curl -fsSL https://ollama.com/install.sh | sh
|
811 |
+
# fetch 模型
|
812 |
+
ollama pull internlm/internlm3-8b-instruct
|
813 |
+
# install python库
|
814 |
+
pip install ollama
|
815 |
+
```
|
816 |
+
|
817 |
+
inference code,
|
818 |
+
|
819 |
+
```python
|
820 |
+
import ollama
|
821 |
+
|
822 |
+
messages = [
|
823 |
+
{
|
824 |
+
"role": "system",
|
825 |
+
"content": thinking_system_prompt,
|
826 |
+
},
|
827 |
+
{
|
828 |
+
"role": "user",
|
829 |
+
"content": "Given the function\(f(x)=\mathrm{e}^{x}-ax - a^{3}\),\n(1) When \(a = 1\), find the equation of the tangent line to the curve \(y = f(x)\) at the point \((1,f(1))\).\n(2) If \(f(x)\) has a local minimum and the minimum value is less than \(0\), determine the range of values for \(a\)."
|
830 |
+
},
|
831 |
+
]
|
832 |
+
|
833 |
+
stream = ollama.chat(
|
834 |
+
model='internlm/internlm3-8b-instruct',
|
835 |
+
messages=messages,
|
836 |
+
stream=True,
|
837 |
+
)
|
838 |
+
|
839 |
+
for chunk in stream:
|
840 |
+
print(chunk['message']['content'], end='', flush=True)
|
841 |
+
```
|
842 |
+
|
843 |
+
|
844 |
+
####
|
845 |
|
846 |
##### vLLM 推理
|
847 |
|
|
|
849 |
|
850 |
```python
|
851 |
git clone https://github.com/RunningLeon/vllm.git
|
852 |
+
# and then follow https://docs.vllm.ai/en/latest/getting_started/installation/gpu/index.html#build-wheel-from-source to install
|
853 |
+
cd vllm
|
854 |
+
python use_existing_torch.py
|
855 |
+
pip install -r requirements-build.txt
|
856 |
+
pip install -e . --no-build-isolatio
|
857 |
```
|
858 |
|
859 |
推理代码
|
|
|
903 |
archivePrefix={arXiv},
|
904 |
primaryClass={cs.CL}
|
905 |
}
|
906 |
+
```
|