Skip to content

Commit d25262c

Browse files
edgararuizedgararuiz
authored
Merge pull request #120 from rainoffallingstar/main
Add support for openai-compatible models from 01AI, moonshot, Qwen,and GLM
2 parents 1b2fc30 + a2774d5 commit d25262c

File tree

4 files changed

+132
-0
lines changed

4 files changed

+132
-0
lines changed

‎inst/configs/glm4.yml

+33
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
default:
2+
prompt: |
3+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
4+
provider: OpenAI - Chat Completions
5+
path: https://open.bigmodel.cn/api/paas/v4/chat/completions
6+
label: glm-4-0520 (GLM)
7+
model: glm-4-0520
8+
max_data_files: 0
9+
max_data_frames: 0
10+
include_doc_contents: FALSE
11+
include_history: TRUE
12+
system_msg: You are a helpful coding assistant
13+
model_arguments:
14+
temperature: 0.03
15+
max_tokens: 1000
16+
stream: TRUE
17+
chat:
18+
prompt: |
19+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
20+
For code output, use RMarkdown code chunks
21+
Avoid all code chunk options
22+
console:
23+
prompt: |
24+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
25+
For any line that is not code, prefix with a: #
26+
Keep each line of explanations to no more than 80 characters
27+
DO NOT use Markdown for the code
28+
script:
29+
prompt: |
30+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
31+
For any line that is not code, prefix with a: #
32+
Keep each line of explanations to no more than 80 characters
33+
DO NOT use Markdown for the code

‎inst/configs/moonshot8k.yml

+33
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
default:
2+
prompt: |
3+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
4+
provider: OpenAI - Chat Completions
5+
path: https://api.moonshot.cn/v1/chat/completions
6+
label: moonshot-v1-8k (Moonshot AI)
7+
model: moonshot-v1-8k
8+
max_data_files: 0
9+
max_data_frames: 0
10+
include_doc_contents: FALSE
11+
include_history: TRUE
12+
system_msg: You are a helpful coding assistant
13+
model_arguments:
14+
temperature: 0.03
15+
max_tokens: 1000
16+
stream: TRUE
17+
chat:
18+
prompt: |
19+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
20+
For code output, use RMarkdown code chunks
21+
Avoid all code chunk options
22+
console:
23+
prompt: |
24+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
25+
For any line that is not code, prefix with a: #
26+
Keep each line of explanations to no more than 80 characters
27+
DO NOT use Markdown for the code
28+
script:
29+
prompt: |
30+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
31+
For any line that is not code, prefix with a: #
32+
Keep each line of explanations to no more than 80 characters
33+
DO NOT use Markdown for the code

‎inst/configs/qwen.yml

+33
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
default:
2+
prompt: |
3+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
4+
provider: OpenAI - Chat Completions
5+
path: https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions
6+
label: qwen-turbo (alibaba)
7+
model: qwen-turbo
8+
max_data_files: 0
9+
max_data_frames: 0
10+
include_doc_contents: FALSE
11+
include_history: TRUE
12+
system_msg: You are a helpful coding assistant
13+
model_arguments:
14+
temperature: 0.03
15+
max_tokens: 1000
16+
stream: TRUE
17+
chat:
18+
prompt: |
19+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
20+
For code output, use RMarkdown code chunks
21+
Avoid all code chunk options
22+
console:
23+
prompt: |
24+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
25+
For any line that is not code, prefix with a: #
26+
Keep each line of explanations to no more than 80 characters
27+
DO NOT use Markdown for the code
28+
script:
29+
prompt: |
30+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
31+
For any line that is not code, prefix with a: #
32+
Keep each line of explanations to no more than 80 characters
33+
DO NOT use Markdown for the code

‎inst/configs/yi.yml

+33
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
default:
2+
prompt: |
3+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
4+
provider: OpenAI - Chat Completions
5+
path: https://api.lingyiwanwu.com/v1/chat/completions
6+
label: yi-34b-chat-0205 (01.AI)
7+
model: yi-34b-chat-0205
8+
max_data_files: 0
9+
max_data_frames: 0
10+
include_doc_contents: FALSE
11+
include_history: TRUE
12+
system_msg: You are a helpful coding assistant
13+
model_arguments:
14+
temperature: 0.03
15+
max_tokens: 1000
16+
stream: TRUE
17+
chat:
18+
prompt: |
19+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
20+
For code output, use RMarkdown code chunks
21+
Avoid all code chunk options
22+
console:
23+
prompt: |
24+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
25+
For any line that is not code, prefix with a: #
26+
Keep each line of explanations to no more than 80 characters
27+
DO NOT use Markdown for the code
28+
script:
29+
prompt: |
30+
{readLines(system.file('prompt/base.txt', package = 'chattr'))}
31+
For any line that is not code, prefix with a: #
32+
Keep each line of explanations to no more than 80 characters
33+
DO NOT use Markdown for the code

0 commit comments

Comments
 (0)