1c20137b0e
support azure deployment name
2024-07-05 19:59:45 +08:00
0aa807df19
feat: remove empty memoryPrompt in ChatMessages
2024-05-16 14:41:18 +08:00
c10447df79
feat: 1)upload image with type 'heic' 2)change the empty message to ';' for models 3)
2024-05-13 16:24:15 +08:00
b3e856df1d
feat: fix 1)the property named 'role' of the first message must be 'user' 2)if default summarize model 'gpt-3.5-turbo' is blocked, use currentModel instead 3)if apiurl&apikey set by location, useCustomConfig would be opened
2024-05-06 19:26:39 +08:00
aa084ea09a
add timezone in system prompts
2024-04-12 23:07:29 +08:00
79f342439a
feat: Solve the problem of using openai interface protocol for user-defined claude model & add some famous webdav endpoints
2024-04-09 20:49:51 +08:00
264da6798c
feat: remove duplicate Input Template
2024-04-08 18:06:17 +08:00
86b5c55855
feat: roles must alternate between user and assistant in claude, so add a fake assistant message between two user messages
2024-04-07 18:02:31 +08:00
768decde93
feat: parse response message
2024-04-07 15:20:27 +08:00
69b079c86e
feat: dev done
2024-04-07 11:32:57 +08:00
e38b527ac2
Merge pull request #3205 from H0llyW00dzZ/summarizelogic
...
Refactor Summarize Logic
2024-03-28 15:19:32 +08:00
e2da3406d2
Add vision support ( #4076 )
2024-02-20 18:04:32 +08:00
bca74241e6
fix: fix gemini issue when using app ( #4013 )
...
* chore: update path
* fix: fix google auth logic
* fix: not using header authorization for google api
* chore: revert to allow stream
2024-02-07 13:17:11 +08:00
9d5801fb5f
fix: avoiding not operation for custom models ( #4010 )
2024-02-07 10:31:49 +08:00
a5517a1a51
Improve Default System Template ( #3996 )
...
* Feat [UI/UX] [Constant] [DEFAULT System Template] replace hardcoded
- [+] feat(constant.ts): replace hardcoded OpenAI with dynamic ServiceProvider variable in DEFAULT_SYSTEM_TEMPLATE
* Improve [UI/UX] [Chat] "fillTemplateWith"
- [+] feat(chat.ts): add DEFAULT_MODELS to modelConfig
- [+] fix(chat.ts): replace replaceAll with regex in output string replacement
- [+] refactor(chat.ts): use const instead of let for cutoff variable
2024-02-05 13:35:51 +08:00
3ba598633c
Non -GPT model disable system prompt ( #3684 )
2023-12-28 23:52:45 +08:00
199f29e63c
chore: auto concat messages
2023-12-25 04:33:47 +08:00
ae19a0dc5f
chroe: update model name
2023-12-24 04:22:12 +08:00
778e88cb56
chore: resolve conflict
2023-12-24 02:15:30 +08:00
ef9e86b50d
Fix UI/UX Page Chats [Memory Prompt] [Stored Local Storage]
...
- [+] fix(chat.ts): update the memory prompt in onFinish callback
- [+] feat(chat.ts): update the current session with lastSummarizeIndex and memoryPrompt
2023-11-25 05:03:34 +07:00
544bab0fe2
Refactor Summarize Logic
...
[+] chore(chat.ts): remove unnecessary comment and refactor variable name
[+] feat(chat.ts): add stream: false to config object
2023-11-09 20:56:45 +07:00
fbc0236748
fix : #3174 should prompt to confirm to delete chat
2023-11-09 02:03:05 +08:00
836bf836d3
fix : #3152 system prompt should be injected
2023-11-07 23:54:30 +08:00
cdf0311d27
feat: add claude and bard
2023-11-07 23:22:11 +08:00
d2d615c84a
Refactor KnowledgeCutoff
...
[+] fix(constant.ts): update DEFAULT_SYSTEM_TEMPLATE to include knowledgeCutoff and time variables
[+] feat(chat.ts): add support for injecting system prompts based on model configuration
2023-11-07 06:42:55 +07:00
5610f423d0
feat: add multi-model support
2023-10-30 02:07:11 +08:00
ad5093ce05
Change log
...
- config.ts
- line 72 remove type confirmation of x as x always has type 'number'
- line 135 remove redundant local variable
- chat.ts
- delete unused import
2023-10-04 17:40:23 +09:00
c73a91a0f5
fixup: fix type errors
2023-09-11 00:24:05 +08:00
38f6956e71
feat: close #2754 add import/export to file
2023-09-11 00:20:23 +08:00
3bd76b9156
feat: close #2580 only use 3.5 to summarize when not using custom models
2023-08-28 00:02:52 +08:00
20a508e2d6
feat: add autoGenerateTitle option
...
(cherry picked from commit 656ab94a9c4edfee820616b8cfc39f5ee9952a3a)
2023-08-14 20:42:09 +08:00
4ab9141429
fix : #2564 should not clear message when error
2023-08-08 21:24:45 +08:00
769c2f9f49
feat: close #2583 do not summarize with gpt-4
2023-08-08 21:22:41 +08:00
7073fd2f3b
Merge pull request #2392 from imldy/dev_SetInjectSystemPrompts
...
fix: enable `enableInjectSystemPrompts` attribute for old sessions
2023-07-16 14:29:50 +08:00
a9f67a48a1
dev: use current inject configuration
2023-07-15 02:48:47 +08:00
fd058cc693
fix: enable `enableInjectSystemPrompts` attribute for old sessions
2023-07-15 01:32:39 +08:00
c00a63e4c3
fix : #2336 resending message should delete origional messages
2023-07-10 22:59:12 +08:00
1513881eed
fix: 修改是否应该注入System Prompt的判断规则为根据设置项
2023-07-10 18:57:54 +08:00
8e4743e719
feat: #920 migrate id to nanoid
2023-07-09 19:37:42 +08:00
089e3b8946
fix summarize don't send
2023-07-05 14:27:37 +08:00
5f0cda829f
feat: #2144 improve input template
2023-06-26 13:18:59 +08:00
ae1ef3215b
feat: close #628 add chat commands
2023-06-24 23:38:11 +08:00
91d8f9d73e
fix : #2111 make "Attached Messages Count" working
2023-06-24 05:53:39 +08:00
1722f75dcb
fixup
2023-06-24 00:18:27 +08:00
be597a551d
feat: #close 1789 add user input template
2023-06-24 00:07:46 +08:00
db63a5a670
fix : #1771 should not lose chat context when sumindex > n - count
2023-06-18 00:51:16 +08:00
4dd5bf71ea
Update chat.ts
2023-06-15 13:59:38 +08:00
1d42e955fc
fix: updating the array using push in zustand does not actually trigger component updates
2023-06-15 10:55:25 +08:00
e636d486f5
fix: updating the array using push in zustand does not actually trigger component updates
2023-06-15 10:23:01 +08:00
76fdd047e7
feat: new token count function
2023-06-15 00:14:38 +08:00
lloydzhou
Dean-YZG
butterfly
l.tingting
DeanYao
TheRam_
fred-bf
H0llyW00dzZ
reece00
Fred Liang
Yidadaa
KeithHello
imldy
sunls233
Xiaocheng Song
Jiacheng Dong