@lobehub/chat 1.16.6 → 1.16.8
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Potentially problematic release.
This version of @lobehub/chat might be problematic. Click here for more details.
- package/CHANGELOG.md +50 -0
- package/README.md +8 -8
- package/README.zh-CN.md +8 -8
- package/package.json +1 -1
- package/src/components/ModelSelect/index.tsx +1 -9
- package/src/config/modelProviders/ai360.ts +23 -68
- package/src/config/modelProviders/anthropic.ts +39 -2
- package/src/config/modelProviders/baichuan.ts +30 -11
- package/src/config/modelProviders/bedrock.ts +80 -19
- package/src/config/modelProviders/deepseek.ts +11 -2
- package/src/config/modelProviders/fireworksai.ts +6 -2
- package/src/config/modelProviders/google.ts +53 -2
- package/src/config/modelProviders/groq.ts +46 -7
- package/src/config/modelProviders/minimax.ts +2 -0
- package/src/config/modelProviders/openai.ts +102 -1
- package/src/const/discover.ts +1 -0
- package/src/libs/agent-runtime/openai/__snapshots__/index.test.ts.snap +56 -0
- package/src/server/routers/edge/config/__snapshots__/index.test.ts.snap +32 -0
- package/src/server/routers/edge/config/index.test.ts +4 -0
- package/src/types/llm.ts +29 -1
- package/src/utils/format.test.ts +42 -1
- package/src/utils/format.ts +11 -0
- package/src/utils/parseModels.test.ts +8 -0
package/CHANGELOG.md
CHANGED
|
@@ -2,6 +2,56 @@
|
|
|
2
2
|
|
|
3
3
|
# Changelog
|
|
4
4
|
|
|
5
|
+
### [Version 1.16.8](https://github.com/lobehub/lobe-chat/compare/v1.16.7...v1.16.8)
|
|
6
|
+
|
|
7
|
+
<sup>Released on **2024-09-12**</sup>
|
|
8
|
+
|
|
9
|
+
#### 💄 Styles
|
|
10
|
+
|
|
11
|
+
- **misc**: Improve models and add more info for providers and models.
|
|
12
|
+
|
|
13
|
+
<br/>
|
|
14
|
+
|
|
15
|
+
<details>
|
|
16
|
+
<summary><kbd>Improvements and Fixes</kbd></summary>
|
|
17
|
+
|
|
18
|
+
#### Styles
|
|
19
|
+
|
|
20
|
+
- **misc**: Improve models and add more info for providers and models, closes [#3911](https://github.com/lobehub/lobe-chat/issues/3911) ([8a8fc6a](https://github.com/lobehub/lobe-chat/commit/8a8fc6a))
|
|
21
|
+
|
|
22
|
+
</details>
|
|
23
|
+
|
|
24
|
+
<div align="right">
|
|
25
|
+
|
|
26
|
+
[](#readme-top)
|
|
27
|
+
|
|
28
|
+
</div>
|
|
29
|
+
|
|
30
|
+
### [Version 1.16.7](https://github.com/lobehub/lobe-chat/compare/v1.16.6...v1.16.7)
|
|
31
|
+
|
|
32
|
+
<sup>Released on **2024-09-11**</sup>
|
|
33
|
+
|
|
34
|
+
#### 💄 Styles
|
|
35
|
+
|
|
36
|
+
- **misc**: Optimize model token display method.
|
|
37
|
+
|
|
38
|
+
<br/>
|
|
39
|
+
|
|
40
|
+
<details>
|
|
41
|
+
<summary><kbd>Improvements and Fixes</kbd></summary>
|
|
42
|
+
|
|
43
|
+
#### Styles
|
|
44
|
+
|
|
45
|
+
- **misc**: Optimize model token display method, closes [#3697](https://github.com/lobehub/lobe-chat/issues/3697) ([249795c](https://github.com/lobehub/lobe-chat/commit/249795c))
|
|
46
|
+
|
|
47
|
+
</details>
|
|
48
|
+
|
|
49
|
+
<div align="right">
|
|
50
|
+
|
|
51
|
+
[](#readme-top)
|
|
52
|
+
|
|
53
|
+
</div>
|
|
54
|
+
|
|
5
55
|
### [Version 1.16.6](https://github.com/lobehub/lobe-chat/compare/v1.16.5...v1.16.6)
|
|
6
56
|
|
|
7
57
|
<sup>Released on **2024-09-11**</sup>
|
package/README.md
CHANGED
|
@@ -285,14 +285,14 @@ Our marketplace is not just a showcase platform but also a collaborative space.
|
|
|
285
285
|
|
|
286
286
|
<!-- AGENT LIST -->
|
|
287
287
|
|
|
288
|
-
| Recent Submits
|
|
289
|
-
|
|
|
290
|
-
| [
|
|
291
|
-
| [
|
|
292
|
-
| [
|
|
293
|
-
| [
|
|
294
|
-
|
|
295
|
-
> 📊 Total agents: [<kbd>**
|
|
288
|
+
| Recent Submits | Description |
|
|
289
|
+
| ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
|
290
|
+
| [Resume Analysis Expert](https://chat-preview.lobehub.com/market?agent=resume-analyzer)<br/><sup>By **[Pandurangmopgar](https://github.com/Pandurangmopgar)** on **2024-09-11**</sup> | Expert AI assistant for comprehensive resume analysis and job-specific optimization. Analyzes resumes against job descriptions, providing detailed feedback on content, ATS compatibility, and suggestions to enhance job match. Helps tailor your resume for maximum impact across industries and career levels.<br/>`resume` `career` `job-search` `ats` `cv` `analysis` `optimization` `professional-development` `interview-prep` |
|
|
291
|
+
| [AI Agent Generator](https://chat-preview.lobehub.com/market?agent=ai-agent-generator)<br/><sup>By **[xyftw](https://github.com/xyftw)** on **2024-09-10**</sup> | Skilled at creating AI Agent character descriptions that meet the needs.<br/>`ai-agent` `character-creation` |
|
|
292
|
+
| [Desolate Friend](https://chat-preview.lobehub.com/market?agent=meu)<br/><sup>By **[adminewacc](https://github.com/adminewacc)** on **2024-09-10**</sup> | Skilled at comforting and supporting friends<br/>`friendship` `sadness` `support` |
|
|
293
|
+
| [NetMaster](https://chat-preview.lobehub.com/market?agent=net-master)<br/><sup>By **[erhuoyan](https://github.com/erhuoyan)** on **2024-09-10**</sup> | Network Engineer: Professional network topology construction and management<br/>`network-engineer` `network-configuration` `network-management` `network-topology` `network-security` |
|
|
294
|
+
|
|
295
|
+
> 📊 Total agents: [<kbd>**328**</kbd> ](https://github.com/lobehub/lobe-chat-agents)
|
|
296
296
|
|
|
297
297
|
<!-- AGENT LIST -->
|
|
298
298
|
|
package/README.zh-CN.md
CHANGED
|
@@ -273,14 +273,14 @@ LobeChat 的插件生态系统是其核心功能的重要扩展，它极大地
|
|
|
273
273
|
|
|
274
274
|
<!-- AGENT LIST -->
|
|
275
275
|
|
|
276
|
-
| 最近新增
|
|
277
|
-
|
|
|
278
|
-
| [
|
|
279
|
-
| [
|
|
280
|
-
| [
|
|
281
|
-
| [
|
|
282
|
-
|
|
283
|
-
> 📊 Total agents: [<kbd>**
|
|
276
|
+
| 最近新增 | 助手说明 |
|
|
277
|
+
| --------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
|
278
|
+
| [简历分析专家](https://chat-preview.lobehub.com/market?agent=resume-analyzer)<br/><sup>By **[Pandurangmopgar](https://github.com/Pandurangmopgar)** on **2024-09-11**</sup> | 专家级 AI 助手，提供全面的简历分析和职位特定优化。根据职位描述分析简历，提供关于内容、ATS 兼容性和增强职位匹配的详细反馈。帮助您量身定制简历，以在各行业和职业层次上产生最大影响。<br/>`简历` `职业` `求职` `ats` `cv` `分析` `优化` `职业发展` `面试准备` |
|
|
279
|
+
| [AI 代理生成器](https://chat-preview.lobehub.com/market?agent=ai-agent-generator)<br/><sup>By **[xyftw](https://github.com/xyftw)** on **2024-09-10**</sup> | 擅长创建满足需求的 AI 代理角色描述。<br/>`ai-agent` `角色创建` |
|
|
280
|
+
| [孤独的朋友](https://chat-preview.lobehub.com/market?agent=meu)<br/><sup>By **[adminewacc](https://github.com/adminewacc)** on **2024-09-10**</sup> | 擅长安慰和支持朋友<br/>`友谊` `悲伤` `支持` |
|
|
281
|
+
| [NetMaster](https://chat-preview.lobehub.com/market?agent=net-master)<br/><sup>By **[erhuoyan](https://github.com/erhuoyan)** on **2024-09-10**</sup> | 网络工程师：专业网络拓扑搭建与管理<br/>`网络工程师` `网络配置` `网络管理` `网络拓扑` `网络安全` |
|
|
282
|
+
|
|
283
|
+
> 📊 Total agents: [<kbd>**328**</kbd> ](https://github.com/lobehub/lobe-chat-agents)
|
|
284
284
|
|
|
285
285
|
<!-- AGENT LIST -->
|
|
286
286
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@lobehub/chat",
|
|
3
|
-
"version": "1.16.
|
|
3
|
+
"version": "1.16.8",
|
|
4
4
|
"description": "Lobe Chat - an open-source, high-performance chatbot framework that supports speech synthesis, multimodal, and extensible Function Call plugin system. Supports one-click free deployment of your private ChatGPT/LLM web application.",
|
|
5
5
|
"keywords": [
|
|
6
6
|
"framework",
|
|
@@ -10,6 +10,7 @@ import { useTranslation } from 'react-i18next';
|
|
|
10
10
|
import { Center, Flexbox } from 'react-layout-kit';
|
|
11
11
|
|
|
12
12
|
import { ChatModelCard } from '@/types/llm';
|
|
13
|
+
import { formatTokenNumber } from '@/utils/format';
|
|
13
14
|
|
|
14
15
|
const useStyles = createStyles(({ css, token }) => ({
|
|
15
16
|
custom: css`
|
|
@@ -55,15 +56,6 @@ const useStyles = createStyles(({ css, token }) => ({
|
|
|
55
56
|
border-radius: 4px;
|
|
56
57
|
`,
|
|
57
58
|
}));
|
|
58
|
-
const formatTokenNumber = (num: number): string => {
|
|
59
|
-
if (num > 0 && num < 1024) return '1K';
|
|
60
|
-
|
|
61
|
-
let kiloToken = Math.floor(num / 1024);
|
|
62
|
-
if (num >= 128_000 && num < 1_024_000) {
|
|
63
|
-
kiloToken = Math.floor(num / 1000);
|
|
64
|
-
}
|
|
65
|
-
return kiloToken < 1000 ? `${kiloToken}K` : `${Math.floor(kiloToken / 1000)}M`;
|
|
66
|
-
};
|
|
67
59
|
|
|
68
60
|
interface ModelInfoTagsProps extends ChatModelCard {
|
|
69
61
|
directionReverse?: boolean;
|
|
@@ -1,6 +1,5 @@
|
|
|
1
1
|
import { ModelProviderCard } from '@/types/llm';
|
|
2
2
|
|
|
3
|
-
// ref https://ai.360.cn/platform/docs/overview
|
|
4
3
|
const Ai360: ModelProviderCard = {
|
|
5
4
|
chatModels: [
|
|
6
5
|
{
|
|
@@ -9,92 +8,46 @@ const Ai360: ModelProviderCard = {
|
|
|
9
8
|
functionCall: false,
|
|
10
9
|
id: '360gpt2-pro',
|
|
11
10
|
maxOutput: 7000,
|
|
11
|
+
pricing: {
|
|
12
|
+
currency: 'CNY',
|
|
13
|
+
input: 5,
|
|
14
|
+
output: 5,
|
|
15
|
+
},
|
|
12
16
|
tokens: 8192,
|
|
13
17
|
},
|
|
14
18
|
{
|
|
15
19
|
displayName: '360GPT Pro',
|
|
16
|
-
functionCall: false,
|
|
17
20
|
id: '360gpt-pro',
|
|
18
21
|
maxOutput: 7000,
|
|
22
|
+
pricing: {
|
|
23
|
+
currency: 'CNY',
|
|
24
|
+
input: 5,
|
|
25
|
+
output: 5,
|
|
26
|
+
},
|
|
19
27
|
tokens: 8192,
|
|
20
28
|
},
|
|
21
|
-
{
|
|
22
|
-
displayName: '360GPT Pro Perf',
|
|
23
|
-
functionCall: false,
|
|
24
|
-
id: '360gpt-pro-perf',
|
|
25
|
-
maxOutput: 7000,
|
|
26
|
-
tokens: 8192,
|
|
27
|
-
},
|
|
28
|
-
{
|
|
29
|
-
displayName: '360GPT Pro sc202401v3',
|
|
30
|
-
functionCall: false,
|
|
31
|
-
id: '360gpt-pro-sc202401v3',
|
|
32
|
-
maxOutput: 2048,
|
|
33
|
-
tokens: 4096,
|
|
34
|
-
},
|
|
35
|
-
{
|
|
36
|
-
displayName: '360GPT Pro sc202401v2',
|
|
37
|
-
functionCall: false,
|
|
38
|
-
id: '360gpt-pro-sc202401v2',
|
|
39
|
-
maxOutput: 2048,
|
|
40
|
-
tokens: 4096,
|
|
41
|
-
},
|
|
42
|
-
{
|
|
43
|
-
displayName: '360GPT Pro sc202401v1',
|
|
44
|
-
functionCall: false,
|
|
45
|
-
id: '360gpt-pro-sc202401v1',
|
|
46
|
-
maxOutput: 2048,
|
|
47
|
-
tokens: 4096,
|
|
48
|
-
},
|
|
49
|
-
{
|
|
50
|
-
displayName: '360GPT Pro v2.0.3',
|
|
51
|
-
functionCall: false,
|
|
52
|
-
id: '360gpt-pro-v2.0.3',
|
|
53
|
-
maxOutput: 2048,
|
|
54
|
-
tokens: 4096,
|
|
55
|
-
},
|
|
56
29
|
{
|
|
57
30
|
displayName: '360GPT Turbo',
|
|
58
31
|
enabled: true,
|
|
59
|
-
functionCall: false,
|
|
60
32
|
id: '360gpt-turbo',
|
|
61
|
-
maxOutput:
|
|
33
|
+
maxOutput: 7000,
|
|
34
|
+
pricing: {
|
|
35
|
+
currency: 'CNY',
|
|
36
|
+
input: 2,
|
|
37
|
+
output: 2,
|
|
38
|
+
},
|
|
62
39
|
tokens: 8192,
|
|
63
40
|
},
|
|
64
41
|
{
|
|
65
42
|
displayName: '360GPT Turbo Responsibility 8K',
|
|
66
43
|
enabled: true,
|
|
67
|
-
functionCall: false,
|
|
68
44
|
id: '360gpt-turbo-responsibility-8k',
|
|
69
45
|
maxOutput: 2048,
|
|
70
|
-
|
|
71
|
-
|
|
72
|
-
|
|
73
|
-
|
|
74
|
-
|
|
75
|
-
id: '360gpt-turbo-32k-responsibility-240530',
|
|
76
|
-
maxOutput: 32_000,
|
|
77
|
-
tokens: 32_000,
|
|
78
|
-
},
|
|
79
|
-
{
|
|
80
|
-
displayName: '360GPT Turbo 32K Responsibility 240516',
|
|
81
|
-
functionCall: false,
|
|
82
|
-
id: '360gpt-turbo-32k-responsibility-240516',
|
|
83
|
-
maxOutput: 32_000,
|
|
84
|
-
tokens: 32_000,
|
|
85
|
-
},
|
|
86
|
-
{
|
|
87
|
-
displayName: '360GPT_S1_QIYUAN',
|
|
88
|
-
functionCall: false,
|
|
89
|
-
id: '360GPT_S1_QIYUAN',
|
|
90
|
-
maxOutput: 2048,
|
|
91
|
-
tokens: 4096,
|
|
92
|
-
},
|
|
93
|
-
{
|
|
94
|
-
displayName: '360GPT_S2_V9',
|
|
95
|
-
functionCall: false,
|
|
96
|
-
id: '360GPT_S2_V9',
|
|
97
|
-
maxOutput: 7000,
|
|
46
|
+
pricing: {
|
|
47
|
+
currency: 'CNY',
|
|
48
|
+
input: 2,
|
|
49
|
+
output: 2,
|
|
50
|
+
},
|
|
98
51
|
tokens: 8192,
|
|
99
52
|
},
|
|
100
53
|
],
|
|
@@ -102,7 +55,9 @@ const Ai360: ModelProviderCard = {
|
|
|
102
55
|
disableBrowserRequest: true,
|
|
103
56
|
id: 'ai360',
|
|
104
57
|
modelList: { showModelFetcher: true },
|
|
58
|
+
modelsUrl: 'https://ai.360.cn/platform/docs/overview',
|
|
105
59
|
name: '360 AI',
|
|
60
|
+
url: 'https://ai.360.com',
|
|
106
61
|
};
|
|
107
62
|
|
|
108
63
|
export default Ai360;
|
|
@@ -1,6 +1,5 @@
|
|
|
1
1
|
import { ModelProviderCard } from '@/types/llm';
|
|
2
2
|
|
|
3
|
-
// ref https://docs.anthropic.com/claude/docs/models-overview#model-comparison
|
|
4
3
|
const Anthropic: ModelProviderCard = {
|
|
5
4
|
chatModels: [
|
|
6
5
|
{
|
|
@@ -11,6 +10,13 @@ const Anthropic: ModelProviderCard = {
|
|
|
11
10
|
functionCall: true,
|
|
12
11
|
id: 'claude-3-5-sonnet-20240620',
|
|
13
12
|
maxOutput: 8192,
|
|
13
|
+
pricing: {
|
|
14
|
+
cachedInput: 0.3,
|
|
15
|
+
input: 3,
|
|
16
|
+
output: 15,
|
|
17
|
+
writeCacheInput: 3.75,
|
|
18
|
+
},
|
|
19
|
+
releasedAt: '2024-06-20',
|
|
14
20
|
tokens: 200_000,
|
|
15
21
|
vision: true,
|
|
16
22
|
},
|
|
@@ -18,10 +24,14 @@ const Anthropic: ModelProviderCard = {
|
|
|
18
24
|
description:
|
|
19
25
|
'Ideal balance of intelligence and speed for enterprise workloads. Maximum utility at a lower price, dependable, balanced for scaled deployments',
|
|
20
26
|
displayName: 'Claude 3 Sonnet',
|
|
21
|
-
enabled: true,
|
|
22
27
|
functionCall: true,
|
|
23
28
|
id: 'claude-3-sonnet-20240229',
|
|
24
29
|
maxOutput: 4096,
|
|
30
|
+
pricing: {
|
|
31
|
+
input: 3,
|
|
32
|
+
output: 15,
|
|
33
|
+
},
|
|
34
|
+
releasedAt: '2024-02-29',
|
|
25
35
|
tokens: 200_000,
|
|
26
36
|
vision: true,
|
|
27
37
|
},
|
|
@@ -33,6 +43,11 @@ const Anthropic: ModelProviderCard = {
|
|
|
33
43
|
functionCall: true,
|
|
34
44
|
id: 'claude-3-opus-20240229',
|
|
35
45
|
maxOutput: 4096,
|
|
46
|
+
pricing: {
|
|
47
|
+
input: 15,
|
|
48
|
+
output: 75,
|
|
49
|
+
},
|
|
50
|
+
releasedAt: '2024-02-29',
|
|
36
51
|
tokens: 200_000,
|
|
37
52
|
vision: true,
|
|
38
53
|
},
|
|
@@ -44,6 +59,11 @@ const Anthropic: ModelProviderCard = {
|
|
|
44
59
|
functionCall: true,
|
|
45
60
|
id: 'claude-3-haiku-20240307',
|
|
46
61
|
maxOutput: 4096,
|
|
62
|
+
pricing: {
|
|
63
|
+
input: 0.25,
|
|
64
|
+
output: 1.25,
|
|
65
|
+
},
|
|
66
|
+
releasedAt: '2024-03-07',
|
|
47
67
|
tokens: 200_000,
|
|
48
68
|
vision: true,
|
|
49
69
|
},
|
|
@@ -51,27 +71,44 @@ const Anthropic: ModelProviderCard = {
|
|
|
51
71
|
displayName: 'Claude 2.1',
|
|
52
72
|
id: 'claude-2.1',
|
|
53
73
|
maxOutput: 4096,
|
|
74
|
+
pricing: {
|
|
75
|
+
input: 8,
|
|
76
|
+
output: 24,
|
|
77
|
+
},
|
|
78
|
+
releasedAt: '2023-11-21',
|
|
54
79
|
tokens: 200_000,
|
|
55
80
|
},
|
|
56
81
|
{
|
|
57
82
|
displayName: 'Claude 2.0',
|
|
58
83
|
id: 'claude-2.0',
|
|
59
84
|
maxOutput: 4096,
|
|
85
|
+
pricing: {
|
|
86
|
+
input: 8,
|
|
87
|
+
output: 24,
|
|
88
|
+
},
|
|
89
|
+
releasedAt: '2023-07-11',
|
|
60
90
|
tokens: 100_000,
|
|
61
91
|
},
|
|
62
92
|
{
|
|
63
93
|
displayName: 'Claude Instant 1.2',
|
|
64
94
|
id: 'claude-instant-1.2',
|
|
65
95
|
maxOutput: 4096,
|
|
96
|
+
pricing: {
|
|
97
|
+
input: 0.8,
|
|
98
|
+
output: 2.4,
|
|
99
|
+
},
|
|
100
|
+
releasedAt: '2023-08-09',
|
|
66
101
|
tokens: 100_000,
|
|
67
102
|
},
|
|
68
103
|
],
|
|
69
104
|
checkModel: 'claude-3-haiku-20240307',
|
|
70
105
|
id: 'anthropic',
|
|
106
|
+
modelsUrl: 'https://docs.anthropic.com/en/docs/about-claude/models#model-names',
|
|
71
107
|
name: 'Anthropic',
|
|
72
108
|
proxyUrl: {
|
|
73
109
|
placeholder: 'https://api.anthropic.com',
|
|
74
110
|
},
|
|
111
|
+
url: 'https://anthropic.com',
|
|
75
112
|
};
|
|
76
113
|
|
|
77
114
|
export default Anthropic;
|
|
@@ -4,54 +4,73 @@ import { ModelProviderCard } from '@/types/llm';
|
|
|
4
4
|
const Baichuan: ModelProviderCard = {
|
|
5
5
|
chatModels: [
|
|
6
6
|
{
|
|
7
|
-
description:
|
|
7
|
+
description:
|
|
8
|
+
'模型能力国内第一，在知识百科、长文本、生成创作等中文任务上超越国外主流模型。还具备行业领先的多模态能力，多项权威评测基准表现优异。',
|
|
8
9
|
displayName: 'Baichuan 4',
|
|
9
10
|
enabled: true,
|
|
10
11
|
functionCall: true,
|
|
11
12
|
id: 'Baichuan4',
|
|
12
13
|
maxOutput: 4096,
|
|
14
|
+
pricing: {
|
|
15
|
+
currency: 'CNY',
|
|
16
|
+
input: 100,
|
|
17
|
+
output: 100,
|
|
18
|
+
},
|
|
13
19
|
tokens: 32_768,
|
|
14
20
|
},
|
|
15
21
|
{
|
|
16
|
-
description:
|
|
22
|
+
description:
|
|
23
|
+
'针对企业高频场景优化，效果大幅提升，高性价比。相对于Baichuan2模型，内容创作提升20%，知识问答提升17%， 角色扮演能力提升40%。整体效果比GPT3.5更优。',
|
|
17
24
|
displayName: 'Baichuan 3 Turbo',
|
|
18
25
|
enabled: true,
|
|
19
26
|
functionCall: true,
|
|
20
27
|
id: 'Baichuan3-Turbo',
|
|
21
28
|
maxOutput: 8192,
|
|
29
|
+
pricing: {
|
|
30
|
+
currency: 'CNY',
|
|
31
|
+
input: 12,
|
|
32
|
+
output: 12,
|
|
33
|
+
},
|
|
22
34
|
tokens: 32_768,
|
|
23
35
|
},
|
|
24
36
|
{
|
|
25
|
-
description:
|
|
37
|
+
description:
|
|
38
|
+
'具备 128K 超长上下文窗口，针对企业高频场景优化，效果大幅提升，高性价比。相对于Baichuan2模型，内容创作提升20%，知识问答提升17%， 角色扮演能力提升40%。整体效果比GPT3.5更优。',
|
|
26
39
|
displayName: 'Baichuan 3 Turbo 128k',
|
|
27
40
|
enabled: true,
|
|
28
41
|
id: 'Baichuan3-Turbo-128k',
|
|
29
42
|
maxOutput: 4096,
|
|
43
|
+
pricing: {
|
|
44
|
+
currency: 'CNY',
|
|
45
|
+
input: 24,
|
|
46
|
+
output: 24,
|
|
47
|
+
},
|
|
30
48
|
tokens: 128_000,
|
|
31
49
|
},
|
|
32
50
|
{
|
|
33
|
-
description:
|
|
51
|
+
description:
|
|
52
|
+
'采用搜索增强技术实现大模型与领域知识、全网知识的全面链接。支持PDF、Word等多种文档上传及网址输入，信息获取及时、全面，输出结果准确、专业。',
|
|
34
53
|
displayName: 'Baichuan 2 Turbo',
|
|
35
54
|
id: 'Baichuan2-Turbo',
|
|
36
55
|
maxOutput: 8192,
|
|
56
|
+
pricing: {
|
|
57
|
+
currency: 'CNY',
|
|
58
|
+
input: 8,
|
|
59
|
+
output: 8,
|
|
60
|
+
},
|
|
37
61
|
tokens: 32_768,
|
|
38
62
|
},
|
|
39
|
-
{
|
|
40
|
-
description: '具备 192K 超长上下文窗口，采用搜索增强技术实现大模型与领域知识、全网知识的全面链接。支持PDF、Word等多种文档上传及网址输入，信息获取及时、全面，输出结果准确、专业。',
|
|
41
|
-
displayName: 'Baichuan 2 Turbo 192k',
|
|
42
|
-
id: 'Baichuan2-Turbo-192k',
|
|
43
|
-
maxOutput: 2048,
|
|
44
|
-
tokens: 192_000,
|
|
45
|
-
},
|
|
46
63
|
],
|
|
47
64
|
checkModel: 'Baichuan3-Turbo',
|
|
48
65
|
id: 'baichuan',
|
|
49
66
|
modelList: { showModelFetcher: true },
|
|
67
|
+
modelsUrl: 'https://platform.baichuan-ai.com/price',
|
|
50
68
|
name: 'Baichuan',
|
|
51
69
|
smoothing: {
|
|
52
70
|
speed: 2,
|
|
53
71
|
text: true,
|
|
54
72
|
},
|
|
73
|
+
url: 'https://platform.baichuan-ai.com',
|
|
55
74
|
};
|
|
56
75
|
|
|
57
76
|
export default Baichuan;
|