@lobehub/chat 1.94.16 → 1.95.0
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.
- package/CHANGELOG.md +50 -0
- package/README.zh-CN.md +8 -8
- package/changelog/v1.json +18 -0
- package/docs/self-hosting/advanced/online-search.mdx +123 -5
- package/docs/self-hosting/advanced/online-search.zh-CN.mdx +123 -4
- package/locales/ar/models.json +2 -2
- package/locales/bg-BG/models.json +2 -2
- package/locales/de-DE/models.json +2 -2
- package/locales/en-US/models.json +2 -2
- package/locales/es-ES/models.json +2 -2
- package/locales/fa-IR/models.json +2 -2
- package/locales/fr-FR/models.json +2 -2
- package/locales/it-IT/models.json +2 -2
- package/locales/ja-JP/models.json +2 -2
- package/locales/ko-KR/models.json +2 -2
- package/locales/nl-NL/models.json +2 -2
- package/locales/pl-PL/models.json +2 -2
- package/locales/pt-BR/models.json +2 -2
- package/locales/ru-RU/models.json +2 -2
- package/locales/tr-TR/models.json +2 -2
- package/locales/vi-VN/models.json +2 -2
- package/locales/zh-CN/models.json +2 -2
- package/locales/zh-TW/models.json +2 -2
- package/package.json +1 -1
- package/src/server/services/search/impls/anspire/index.ts +132 -0
- package/src/server/services/search/impls/anspire/type.ts +21 -0
- package/src/server/services/search/impls/brave/index.ts +129 -0
- package/src/server/services/search/impls/brave/type.ts +58 -0
- package/src/server/services/search/impls/google/index.ts +129 -0
- package/src/server/services/search/impls/google/type.ts +53 -0
- package/src/server/services/search/impls/index.ts +24 -0
- package/src/server/services/search/impls/kagi/index.ts +111 -0
- package/src/server/services/search/impls/kagi/type.ts +24 -0
package/CHANGELOG.md
CHANGED
|
@@ -2,6 +2,56 @@
|
|
|
2
2
|
|
|
3
3
|
# Changelog
|
|
4
4
|
|
|
5
|
+
## [Version 1.95.0](https://github.com/lobehub/lobe-chat/compare/v1.94.17...v1.95.0)
|
|
6
|
+
|
|
7
|
+
<sup>Released on **2025-06-20**</sup>
|
|
8
|
+
|
|
9
|
+
#### ✨ Features
|
|
10
|
+
|
|
11
|
+
- **misc**: Add Brave & Google PSE & Kagi as build-in Search Provider.
|
|
12
|
+
|
|
13
|
+
<br/>
|
|
14
|
+
|
|
15
|
+
<details>
|
|
16
|
+
<summary><kbd>Improvements and Fixes</kbd></summary>
|
|
17
|
+
|
|
18
|
+
#### What's improved
|
|
19
|
+
|
|
20
|
+
- **misc**: Add Brave & Google PSE & Kagi as build-in Search Provider, closes [#8172](https://github.com/lobehub/lobe-chat/issues/8172) ([16ae521](https://github.com/lobehub/lobe-chat/commit/16ae521))
|
|
21
|
+
|
|
22
|
+
</details>
|
|
23
|
+
|
|
24
|
+
<div align="right">
|
|
25
|
+
|
|
26
|
+
[](#readme-top)
|
|
27
|
+
|
|
28
|
+
</div>
|
|
29
|
+
|
|
30
|
+
### [Version 1.94.17](https://github.com/lobehub/lobe-chat/compare/v1.94.16...v1.94.17)
|
|
31
|
+
|
|
32
|
+
<sup>Released on **2025-06-20**</sup>
|
|
33
|
+
|
|
34
|
+
#### 💄 Styles
|
|
35
|
+
|
|
36
|
+
- **misc**: Update i18n.
|
|
37
|
+
|
|
38
|
+
<br/>
|
|
39
|
+
|
|
40
|
+
<details>
|
|
41
|
+
<summary><kbd>Improvements and Fixes</kbd></summary>
|
|
42
|
+
|
|
43
|
+
#### Styles
|
|
44
|
+
|
|
45
|
+
- **misc**: Update i18n, closes [#8233](https://github.com/lobehub/lobe-chat/issues/8233) ([88c4362](https://github.com/lobehub/lobe-chat/commit/88c4362))
|
|
46
|
+
|
|
47
|
+
</details>
|
|
48
|
+
|
|
49
|
+
<div align="right">
|
|
50
|
+
|
|
51
|
+
[](#readme-top)
|
|
52
|
+
|
|
53
|
+
</div>
|
|
54
|
+
|
|
5
55
|
### [Version 1.94.16](https://github.com/lobehub/lobe-chat/compare/v1.94.15...v1.94.16)
|
|
6
56
|
|
|
7
57
|
<sup>Released on **2025-06-19**</sup>
|
package/README.zh-CN.md
CHANGED
|
@@ -356,14 +356,14 @@ LobeChat 的插件生态系统是其核心功能的重要扩展，它极大地
|
|
|
356
356
|
|
|
357
357
|
<!-- AGENT LIST -->
|
|
358
358
|
|
|
359
|
-
| 最近新增
|
|
360
|
-
|
|
|
361
|
-
| [
|
|
362
|
-
| [
|
|
363
|
-
| [
|
|
364
|
-
| [
|
|
365
|
-
|
|
366
|
-
> 📊 Total agents: [<kbd>**
|
|
359
|
+
| 最近新增 | 描述 |
|
|
360
|
+
| ---------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------- |
|
|
361
|
+
| [海龟汤主持人](https://lobechat.com/discover/assistant/lateral-thinking-puzzle)<br/><sup>By **[CSY2022](https://github.com/CSY2022)** on **2025-06-19**</sup> | 一个海龟汤主持人，需要自己提供汤面，汤底与关键点（猜中的判定条件）。<br/>`海龟汤` `推理` `互动` `谜题` `角色扮演` |
|
|
362
|
+
| [美食评论员🍟](https://lobechat.com/discover/assistant/food-reviewer)<br/><sup>By **[renhai-lab](https://github.com/renhai-lab)** on **2025-06-17**</sup> | 美食评价专家<br/>`美食` `评价` `写作` |
|
|
363
|
+
| [学术写作助手](https://lobechat.com/discover/assistant/academic-writing-assistant)<br/><sup>By **[swarfte](https://github.com/swarfte)** on **2025-06-17**</sup> | 专业的学术研究论文写作和正式文档编写专家<br/>`学术写作` `研究` `正式风格` |
|
|
364
|
+
| [Minecraft 资深开发者](https://lobechat.com/discover/assistant/java-development)<br/><sup>By **[iamyuuk](https://github.com/iamyuuk)** on **2025-06-17**</sup> | 擅长高级 Java 开发及 Minecraft 开发<br/>`开发` `编程` `minecraft` `java` |
|
|
365
|
+
|
|
366
|
+
> 📊 Total agents: [<kbd>**505**</kbd> ](https://lobechat.com/discover/assistants)
|
|
367
367
|
|
|
368
368
|
<!-- AGENT LIST -->
|
|
369
369
|
|
package/changelog/v1.json
CHANGED
|
@@ -1,4 +1,22 @@
|
|
|
1
1
|
[
|
|
2
|
+
{
|
|
3
|
+
"children": {
|
|
4
|
+
"features": [
|
|
5
|
+
"Add Brave & Google PSE & Kagi as build-in Search Provider."
|
|
6
|
+
]
|
|
7
|
+
},
|
|
8
|
+
"date": "2025-06-20",
|
|
9
|
+
"version": "1.95.0"
|
|
10
|
+
},
|
|
11
|
+
{
|
|
12
|
+
"children": {
|
|
13
|
+
"improvements": [
|
|
14
|
+
"Update i18n."
|
|
15
|
+
]
|
|
16
|
+
},
|
|
17
|
+
"date": "2025-06-20",
|
|
18
|
+
"version": "1.94.17"
|
|
19
|
+
},
|
|
2
20
|
{
|
|
3
21
|
"children": {
|
|
4
22
|
"fixes": [
|
|
@@ -14,16 +14,134 @@ tags:
|
|
|
14
14
|
|
|
15
15
|
# Configuring Online Search Functionality
|
|
16
16
|
|
|
17
|
-
LobeChat supports configuring
|
|
17
|
+
LobeChat supports configuring **web search functionality** for AI, enabling it to retrieve real-time information from the internet to provide more accurate and up-to-date responses. Web search supports multiple search engine providers, including [SearXNG](https://github.com/searxng/searxng), [Search1API](https://www.search1api.com), [Google](https://programmablesearchengine.google.com), and [Brave](https://brave.com/search/api), among others.
|
|
18
18
|
|
|
19
|
-
<Callout type=
|
|
20
|
-
|
|
21
|
-
instances. By configuring SearXNG, LobeChat enables AI to retrieve the latest internet
|
|
22
|
-
information, allowing it to answer time-sensitive questions and provide up-to-date news.
|
|
19
|
+
<Callout type="info">
|
|
20
|
+
Web search allows AI to access time-sensitive content, such as the latest news, technology trends, or product information. You can deploy the open-source SearXNG yourself, or choose to integrate mainstream search services like Search1API, Google, Brave, etc., combining them freely based on your use case.
|
|
23
21
|
</Callout>
|
|
24
22
|
|
|
23
|
+
By setting the search service environment variable `SEARCH_PROVIDERS` and the corresponding API Keys, LobeChat will query multiple sources and return the results. You can also configure crawler service environment variables such as `CRAWLER_IMPLS` (e.g., `browserless`, `firecrawl`, `tavily`, etc.) to extract webpage content, enhancing the capability of search + reading.
|
|
24
|
+
|
|
25
25
|
# Core Environment Variables
|
|
26
26
|
|
|
27
|
+
## `CRAWLER_IMPLS`
|
|
28
|
+
|
|
29
|
+
Configure available web crawlers for structured extraction of webpage content.
|
|
30
|
+
|
|
31
|
+
```env
|
|
32
|
+
CRAWLER_IMPLS="native,search1api"
|
|
33
|
+
```
|
|
34
|
+
|
|
35
|
+
Supported crawler types are listed below:
|
|
36
|
+
|
|
37
|
+
| Value | Description | Environment Variable |
|
|
38
|
+
| ------------- | ------------------------------------------------------------------------------------------------------------------- | -------------------------- |
|
|
39
|
+
| `browserless` | Headless browser crawler based on [Browserless](https://www.browserless.io/), suitable for rendering complex pages. | `BROWSERLESS_TOKEN` |
|
|
40
|
+
| `exa` | Crawler capabilities provided by [Exa](https://exa.ai/), API required. | `EXA_API_KEY` |
|
|
41
|
+
| `firecrawl` | [Firecrawl](https://firecrawl.dev/) headless browser API, ideal for modern websites. | `FIRECRAWL_API_KEY` |
|
|
42
|
+
| `jina` | Crawler service from [Jina AI](https://jina.ai/), supports fast content summarization. | `JINA_READER_API_KEY` |
|
|
43
|
+
| `native` | Built-in general-purpose crawler for standard web structures. | |
|
|
44
|
+
| `search1api` | Page crawling capabilities from [Search1API](https://www.search1api.com), great for structured content extraction. | `SEARCH1API_CRAWL_API_KEY` |
|
|
45
|
+
| `tavily` | Web scraping and summarization API from [Tavily](https://www.tavily.com/). | `TAVILY_API_KEY` |
|
|
46
|
+
|
|
47
|
+
> 💡 Setting multiple crawlers increases success rate; the system will try different ones based on priority.
|
|
48
|
+
|
|
49
|
+
---
|
|
50
|
+
|
|
51
|
+
## `SEARCH_PROVIDERS`
|
|
52
|
+
|
|
53
|
+
Configure which search engine providers to use for web search.
|
|
54
|
+
|
|
55
|
+
```env
|
|
56
|
+
SEARCH_PROVIDERS="searxng"
|
|
57
|
+
```
|
|
58
|
+
|
|
59
|
+
Supported search engines include:
|
|
60
|
+
|
|
61
|
+
| Value | Description | Environment Variable |
|
|
62
|
+
| ------------ | ---------------------------------------------------------------------------------------- | ------------------------------------------- |
|
|
63
|
+
| `anspire` | Search service provided by [Anspire](https://anspire.ai/). | `ANSPIRE_API_KEY` |
|
|
64
|
+
| `bocha` | Search service from [Bocha](https://open.bochaai.com/). | `BOCHA_API_KEY` |
|
|
65
|
+
| `brave` | [Brave](https://search.brave.com/help/api), a privacy-friendly search source. | `BRAVE_API_KEY` |
|
|
66
|
+
| `exa` | [Exa](https://exa.ai/), a search API designed for AI. | `EXA_API_KEY` |
|
|
67
|
+
| `firecrawl` | Search capabilities via [Firecrawl](https://firecrawl.dev/). | `FIRECRAWL_API_KEY` |
|
|
68
|
+
| `google` | Uses [Google Programmable Search Engine](https://programmablesearchengine.google.com/). | `GOOGLE_PSE_API_KEY` `GOOGLE_PSE_ENGINE_ID` |
|
|
69
|
+
| `jina` | Semantic search provided by [Jina AI](https://jina.ai/). | `JINA_READER_API_KEY` |
|
|
70
|
+
| `kagi` | Premium search API by [Kagi](https://kagi.com/), requires a subscription key. | `KAGI_API_KEY` |
|
|
71
|
+
| `search1api` | Aggregated search capabilities from [Search1API](https://www.search1api.com). | `SEARCH1API_CRAWL_API_KEY` |
|
|
72
|
+
| `searxng` | Use a self-hosted or public [SearXNG](https://searx.space/) instance. | `SEARXNG_URL` |
|
|
73
|
+
| `tavily` | [Tavily](https://www.tavily.com/), offers fast web summaries and answers. | `TAVILY_API_KEY` |
|
|
74
|
+
|
|
75
|
+
> ⚠️ Some search providers require you to apply for an API Key and configure it in your `.env` file.
|
|
76
|
+
|
|
77
|
+
---
|
|
78
|
+
|
|
79
|
+
## `BROWSERLESS_URL`
|
|
80
|
+
|
|
81
|
+
Specifies the API endpoint for [Browserless](https://www.browserless.io/), used for web crawling tasks. Browserless is a browser automation platform based on Headless Chrome, ideal for rendering dynamic pages.
|
|
82
|
+
|
|
83
|
+
```env
|
|
84
|
+
BROWSERLESS_URL=https://chrome.browserless.io
|
|
85
|
+
```
|
|
86
|
+
|
|
87
|
+
> 📌 Usually used together with `CRAWLER_IMPLS=browserless`.
|
|
88
|
+
|
|
89
|
+
---
|
|
90
|
+
|
|
91
|
+
## `GOOGLE_PSE_ENGINE_ID`
|
|
92
|
+
|
|
93
|
+
Configure the Search Engine ID for Google Programmable Search Engine (Google PSE), used to restrict the search scope. Must be used alongside `GOOGLE_PSE_API_KEY`.
|
|
94
|
+
|
|
95
|
+
```env
|
|
96
|
+
GOOGLE_PSE_ENGINE_ID=your-google-cx-id
|
|
97
|
+
```
|
|
98
|
+
|
|
99
|
+
> 🔑 How to get it: Visit [programmablesearchengine.google.com](https://programmablesearchengine.google.com/), create a search engine, and obtain the `cx` parameter.
|
|
100
|
+
|
|
101
|
+
---
|
|
102
|
+
|
|
103
|
+
## `FIRECRAWL_URL`
|
|
104
|
+
|
|
105
|
+
Sets the access URL for the [Firecrawl](https://firecrawl.dev/) API, used for web content scraping. Default value:
|
|
106
|
+
|
|
107
|
+
```env
|
|
108
|
+
FIRECRAWL_URL=https://api.firecrawl.dev/v1
|
|
109
|
+
```
|
|
110
|
+
|
|
111
|
+
> ⚙️ Usually does not need to be changed unless you’re using a self-hosted version or a proxy service.
|
|
112
|
+
|
|
113
|
+
---
|
|
114
|
+
|
|
115
|
+
## `TAVILY_SEARCH_DEPTH`
|
|
116
|
+
|
|
117
|
+
Configure the result depth for [Tavily](https://www.tavily.com/) searches.
|
|
118
|
+
|
|
119
|
+
```env
|
|
120
|
+
TAVILY_SEARCH_DEPTH=basic
|
|
121
|
+
```
|
|
122
|
+
|
|
123
|
+
Supported values:
|
|
124
|
+
|
|
125
|
+
* `basic`: Fast search, returns brief results;
|
|
126
|
+
* `advanced`: Deep search, returns more context and web page details.
|
|
127
|
+
|
|
128
|
+
---
|
|
129
|
+
|
|
130
|
+
## `TAVILY_EXTRACT_DEPTH`
|
|
131
|
+
|
|
132
|
+
Configure how deeply Tavily extracts content from web pages.
|
|
133
|
+
|
|
134
|
+
```env
|
|
135
|
+
TAVILY_EXTRACT_DEPTH=basic
|
|
136
|
+
```
|
|
137
|
+
|
|
138
|
+
Supported values:
|
|
139
|
+
|
|
140
|
+
* `basic`: Extracts basic info like title and content summary;
|
|
141
|
+
* `advanced`: Extracts structured data, lists, charts, and more from web pages.
|
|
142
|
+
|
|
143
|
+
---
|
|
144
|
+
|
|
27
145
|
## `SEARXNG_URL`
|
|
28
146
|
|
|
29
147
|
The URL of the SearXNG instance, which is a necessary configuration to enable the online search functionality. For example:
|
|
@@ -10,15 +10,134 @@ tags:
|
|
|
10
10
|
|
|
11
11
|
# 配置联网搜索功能
|
|
12
12
|
|
|
13
|
-
LobeChat 支持为 AI
|
|
13
|
+
LobeChat 支持为 AI 配置**联网搜索功能**，使其能够实时获取互联网信息，从而提供更准确、最新的回答。联网搜索支持多个搜索引擎提供商，包括 [SearXNG](https://github.com/searxng/searxng)、[Search1API](https://www.search1api.com)、[Google](https://programmablesearchengine.google.com)、[Brave](https://brave.com/search/api) 等。
|
|
14
14
|
|
|
15
|
-
<Callout type=
|
|
16
|
-
SearXNG
|
|
17
|
-
可以让 AI 获取最新的互联网信息，从而回答时效性问题、提供最新资讯。
|
|
15
|
+
<Callout type="info">
|
|
16
|
+
联网搜索可以让 AI 获取时效性内容，如最新新闻、技术动态或产品信息。你可以使用开源的 SearXNG 自行部署，也可以选择集成主流搜索引擎服务，如 Search1API、Google、Brave 等，根据你的使用场景自由组合。
|
|
18
17
|
</Callout>
|
|
19
18
|
|
|
19
|
+
通过设置搜索服务环境变量 `SEARCH_PROVIDERS` 和对应的 API Key，LobeChat 将在多个搜索源中查询并返回结果。你还可以搭配配置爬虫服务环境变量 `CRAWLER_IMPLS`（如 `browserless`、`firecrawl`、`tavily` 等）以提取网页内容，实现搜索+阅读的增强能力。
|
|
20
|
+
|
|
20
21
|
# 核心环境变量
|
|
21
22
|
|
|
23
|
+
## `CRAWLER_IMPLS`
|
|
24
|
+
|
|
25
|
+
配置可用的网页爬虫，用于对网页进行结构化内容提取。
|
|
26
|
+
|
|
27
|
+
```env
|
|
28
|
+
CRAWLER_IMPLS="native,search1api"
|
|
29
|
+
```
|
|
30
|
+
|
|
31
|
+
支持的爬虫类型如下：
|
|
32
|
+
|
|
33
|
+
| 值 | 说明 | 环境变量 |
|
|
34
|
+
| ------------- | -------------------------------------------------------------------------------------- | -------------------------- |
|
|
35
|
+
| `browserless` | 基于 [Browserless](https://www.browserless.io/) 的无头浏览器爬虫，适合渲染复杂页面。 | `BROWSERLESS_TOKEN` |
|
|
36
|
+
| `exa` | 使用 [Exa](https://exa.ai/) 提供的爬虫能力，需申请 API。 | `EXA_API_KEY` |
|
|
37
|
+
| `firecrawl` | [Firecrawl](https://firecrawl.dev/) 无头浏览器 API，适合现代网站抓取。 | `FIRECRAWL_API_KEY` |
|
|
38
|
+
| `jina` | 使用 [Jina AI](https://jina.ai/) 的爬虫服务，支持快速提取摘要信息。 | `JINA_READER_API_KEY` |
|
|
39
|
+
| `native` | 内置通用爬虫，适用于标准网页结构。 | |
|
|
40
|
+
| `search1api` | 利用 [Search1API](https://www.search1api.com) 提供的页面抓取能力，适合结构化内容提取。 | `SEARCH1API_CRAWL_API_KEY` |
|
|
41
|
+
| `tavily` | 使用 [Tavily](https://www.tavily.com/) 的网页抓取与摘要 API。 | `TAVILY_API_KEY` |
|
|
42
|
+
|
|
43
|
+
> 💡 设置多个爬虫可提升成功率，系统将根据优先级尝试不同爬虫。
|
|
44
|
+
|
|
45
|
+
---
|
|
46
|
+
|
|
47
|
+
## `SEARCH_PROVIDERS`
|
|
48
|
+
|
|
49
|
+
配置联网搜索使用的搜索引擎提供商。
|
|
50
|
+
|
|
51
|
+
```env
|
|
52
|
+
SEARCH_PROVIDERS="searxng"
|
|
53
|
+
```
|
|
54
|
+
|
|
55
|
+
支持的搜索引擎如下：
|
|
56
|
+
|
|
57
|
+
| 值 | 说明 | 环境变量 |
|
|
58
|
+
| ------------ | ---------------------------------------------------------------------------------------- | ------------------------------------------- |
|
|
59
|
+
| `anspire` | 基于 [Anspire（安思派）](https://anspire.ai/) 提供的搜索服务。 | `ANSPIRE_API_KEY` |
|
|
60
|
+
| `bocha` | 基于 [Bocha（博查）](https://open.bochaai.com/) 提供的搜索服务。 | `BOCHA_API_KEY` |
|
|
61
|
+
| `brave` | [Brave](https://search.brave.com/help/api)，隐私友好的搜索源。 | `BRAVE_API_KEY` |
|
|
62
|
+
| `exa` | [Exa](https://exa.ai/)，面向 AI 的搜索 API。 | `EXA_API_KEY` |
|
|
63
|
+
| `firecrawl` | 支持 [Firecrawl](https://firecrawl.dev/) 提供的搜索服务。 | `FIRECRAWL_API_KEY` |
|
|
64
|
+
| `google` | 使用 [Google Programmable Search Engine](https://programmablesearchengine.google.com/)。 | `GOOGLE_PSE_API_KEY` `GOOGLE_PSE_ENGINE_ID` |
|
|
65
|
+
| `jina` | 使用 [Jina AI](https://jina.ai/) 提供的语义搜索服务。 | `JINA_READER_API_KEY` |
|
|
66
|
+
| `kagi` | [Kagi](https://kagi.com/) 提供的高级搜索 API，需订阅 Key。 | `KAGI_API_KEY` |
|
|
67
|
+
| `search1api` | 使用 [Search1API](https://www.search1api.com) 聚合搜索能力。 | `SEARCH1API_CRAWL_API_KEY` |
|
|
68
|
+
| `searxng` | 使用自托管或公共 [SearXNG](https://searx.space/) 实例。 | `SEARXNG_URL` |
|
|
69
|
+
| `tavily` | [Tavily](https://www.tavily.com/)，快速网页摘要与答案返回。 | `TAVILY_API_KEY` |
|
|
70
|
+
|
|
71
|
+
> ⚠️ 某些搜索提供商需要单独申请 API Key，并在 `.env` 中配置相关凭证。
|
|
72
|
+
|
|
73
|
+
---
|
|
74
|
+
|
|
75
|
+
## `BROWSERLESS_URL`
|
|
76
|
+
|
|
77
|
+
指定 [Browserless](https://www.browserless.io/) 服务的 API 地址，用于执行网页爬取任务。Browserless 是一个基于无头浏览器（Headless Chrome）的浏览器自动化平台，适合处理需要渲染的动态页面。
|
|
78
|
+
|
|
79
|
+
```env
|
|
80
|
+
BROWSERLESS_URL=https://chrome.browserless.io
|
|
81
|
+
```
|
|
82
|
+
|
|
83
|
+
> 📌 通常需要搭配 `CRAWLER_IMPLS=browserless` 启用。
|
|
84
|
+
|
|
85
|
+
---
|
|
86
|
+
|
|
87
|
+
## `GOOGLE_PSE_ENGINE_ID`
|
|
88
|
+
|
|
89
|
+
配置 Google Programmable Search Engine（Google PSE）的搜索引擎 ID，用于限定搜索范围。需配合 `GOOGLE_PSE_API_KEY` 一起使用。
|
|
90
|
+
|
|
91
|
+
```env
|
|
92
|
+
GOOGLE_PSE_ENGINE_ID=your-google-cx-id
|
|
93
|
+
```
|
|
94
|
+
|
|
95
|
+
> 🔑 获取方式：访问 [programmablesearchengine.google.com](https://programmablesearchengine.google.com/)，创建搜索引擎后获取 `cx` 参数值。
|
|
96
|
+
|
|
97
|
+
---
|
|
98
|
+
|
|
99
|
+
## `FIRECRAWL_URL`
|
|
100
|
+
|
|
101
|
+
设置 [Firecrawl](https://firecrawl.dev/) API 的访问地址。用于网页内容抓取，默认值如下：
|
|
102
|
+
|
|
103
|
+
```env
|
|
104
|
+
FIRECRAWL_URL=https://api.firecrawl.dev/v1
|
|
105
|
+
```
|
|
106
|
+
|
|
107
|
+
> ⚙️ 一般无需修改，除非你使用的是自托管版本或代理服务。
|
|
108
|
+
|
|
109
|
+
---
|
|
110
|
+
|
|
111
|
+
## `TAVILY_SEARCH_DEPTH`
|
|
112
|
+
|
|
113
|
+
配置 [Tavily](https://www.tavily.com/) 搜索的结果深度。
|
|
114
|
+
|
|
115
|
+
```env
|
|
116
|
+
TAVILY_SEARCH_DEPTH=basic
|
|
117
|
+
```
|
|
118
|
+
|
|
119
|
+
支持的值：
|
|
120
|
+
|
|
121
|
+
* `basic`: 快速搜索，返回简要结果；
|
|
122
|
+
* `advanced`: 深度搜索，返回更多上下文和网页信息。
|
|
123
|
+
|
|
124
|
+
---
|
|
125
|
+
|
|
126
|
+
## `TAVILY_EXTRACT_DEPTH`
|
|
127
|
+
|
|
128
|
+
配置 Tavily 在抓取网页内容时的提取深度。
|
|
129
|
+
|
|
130
|
+
```env
|
|
131
|
+
TAVILY_EXTRACT_DEPTH=basic
|
|
132
|
+
```
|
|
133
|
+
|
|
134
|
+
支持的值：
|
|
135
|
+
|
|
136
|
+
* `basic`: 提取标题、正文摘要等基础信息；
|
|
137
|
+
* `advanced`: 提取网页的结构化信息、列表、图表等更多内容。
|
|
138
|
+
|
|
139
|
+
---
|
|
140
|
+
|
|
22
141
|
## `SEARXNG_URL`
|
|
23
142
|
|
|
24
143
|
SearXNG 实例的 URL 地址，这是启用联网搜索功能的必要配置。例如：
|
package/locales/ar/models.json
CHANGED
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash هو النموذج الرائد الأكثر تقدمًا من Google، مصمم للاستدلال المتقدم، الترميز، المهام الرياضية والعلمية. يحتوي على قدرة \"التفكير\" المدمجة، مما يمكّنه من تقديم استجابات بدقة أعلى ومعالجة سياقات أكثر تفصيلاً.\n\nملاحظة: يحتوي هذا النموذج على نوعين: التفكير وغير التفكير. تختلف تسعير الإخراج بشكل ملحوظ بناءً على ما إذا كانت قدرة التفكير مفعلة. إذا اخترت النوع القياسي (بدون لاحقة \" :thinking \")، سيتجنب النموذج بشكل صريح توليد رموز التفكير.\n\nلاستغلال قدرة التفكير واستقبال رموز التفكير، يجب عليك اختيار النوع \" :thinking \"، مما سيؤدي إلى تسعير إخراج تفكير أعلى.\n\nبالإضافة إلى ذلك، يمكن تكوين Gemini 2.5 Flash من خلال معلمة \"الحد الأقصى لعدد رموز الاستدلال\"، كما هو موضح في الوثائق (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "معاينة Gemini 2.5 Pro هي أحدث نموذج تفكيري من Google، قادر على استنتاج المشكلات المعقدة في مجالات البرمجة والرياضيات والعلوم والتكنولوجيا والهندسة والرياضيات (STEM)، بالإضافة إلى استخدام سياق طويل لتحليل مجموعات البيانات الكبيرة، وقواعد الشيفرة، والوثائق."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "يقدم Gemini 1.5 Flash قدرات معالجة متعددة الوسائط محسّنة، مناسبة لمجموعة متنوعة من سيناريوهات المهام المعقدة."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash е най-напредналият основен модел на Google, проектиран за напреднали разсъждения, кодиране, математика и научни задачи. Той включва вградена способност за \"мислене\", което му позволява да предоставя отговори с по-висока точност и детайлна обработка на контекста.\n\nЗабележка: Този модел има два варианта: с мислене и без мислене. Цените на изхода значително варират в зависимост от активирането на способността за мислене. Ако изберете стандартния вариант (без суфикс \":thinking\"), моделът ще избягва генерирането на токени за мислене.\n\nЗа да се възползвате от способността за мислене и да получите токени за мислене, трябва да изберете варианта \":thinking\", което ще доведе до по-високи цени на изхода за мислене.\n\nОсвен това, Gemini 2.5 Flash може да бъде конфигуриран чрез параметъра \"максимален брой токени за разсъждение\", както е описано в документацията (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro е
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview е най-усъвършенстваният мисловен модел на Google, способен да извършва разсъждения върху сложни проблеми в областта на кодирането, математиката и STEM, както и да анализира големи набори от данни, кодови бази и документи с дълъг контекст."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash предлага оптимизирани мултимодални обработващи способности, подходящи за различни сложни задачи."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash ist Googles fortschrittlichstes Hauptmodell, das für fortgeschrittenes Denken, Codierung, Mathematik und wissenschaftliche Aufgaben entwickelt wurde. Es enthält die eingebaute Fähigkeit zu \"denken\", was es ihm ermöglicht, Antworten mit höherer Genauigkeit und detaillierter Kontextverarbeitung zu liefern.\n\nHinweis: Dieses Modell hat zwei Varianten: Denken und Nicht-Denken. Die Ausgabepreise variieren erheblich, je nachdem, ob die Denkfähigkeit aktiviert ist oder nicht. Wenn Sie die Standardvariante (ohne den Suffix \":thinking\") wählen, wird das Modell ausdrücklich vermeiden, Denk-Tokens zu generieren.\n\nUm die Denkfähigkeit zu nutzen und Denk-Tokens zu erhalten, müssen Sie die \":thinking\"-Variante wählen, was zu höheren Preisen für Denk-Ausgaben führt.\n\nDarüber hinaus kann Gemini 2.5 Flash über den Parameter \"maximale Tokenanzahl für das Denken\" konfiguriert werden, wie in der Dokumentation beschrieben (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro ist Googles fortschrittlichstes
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview ist Googles fortschrittlichstes Denkmodell, das in der Lage ist, komplexe Probleme in den Bereichen Code, Mathematik und MINT zu analysieren sowie große Datensätze, Codebasen und Dokumente mit langem Kontext zu untersuchen."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash bietet optimierte multimodale Verarbeitungsfähigkeiten, die für verschiedene komplexe Aufgabenszenarien geeignet sind."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash is Google's most advanced flagship model, designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in 'thinking' capabilities that allow it to provide responses with higher accuracy and detailed context handling.\n\nNote: This model has two variants: thinking and non-thinking. Output pricing varies significantly based on whether the thinking capability is activated. If you choose the standard variant (without the ':thinking' suffix), the model will explicitly avoid generating thinking tokens.\n\nTo leverage the thinking capability and receive thinking tokens, you must select the ':thinking' variant, which will incur higher thinking output pricing.\n\nAdditionally, Gemini 2.5 Flash can be configured via the 'maximum tokens for reasoning' parameter, as described in the documentation (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro is Google's
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview is Google's most advanced thinking model, capable of reasoning through complex problems in code, mathematics, and STEM fields, as well as analyzing large datasets, codebases, and documents using extended context."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash offers optimized multimodal processing capabilities, suitable for various complex task scenarios."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash es el modelo principal más avanzado de Google, diseñado para razonamiento avanzado, codificación, matemáticas y tareas científicas. Incluye la capacidad de 'pensar' incorporada, lo que le permite proporcionar respuestas con mayor precisión y un manejo más detallado del contexto.\n\nNota: Este modelo tiene dos variantes: con pensamiento y sin pensamiento. La fijación de precios de salida varía significativamente según si la capacidad de pensamiento está activada. Si elige la variante estándar (sin el sufijo ':thinking'), el modelo evitará explícitamente generar tokens de pensamiento.\n\nPara aprovechar la capacidad de pensamiento y recibir tokens de pensamiento, debe elegir la variante ':thinking', lo que resultará en un precio de salida de pensamiento más alto.\n\nAdemás, Gemini 2.5 Flash se puede configurar a través del parámetro 'número máximo de tokens de razonamiento', como se describe en la documentación (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro es el modelo de
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview es el modelo de pensamiento más avanzado de Google, capaz de razonar sobre problemas complejos en código, matemáticas y áreas STEM, así como de analizar grandes conjuntos de datos, bases de código y documentos utilizando contextos extensos."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash ofrece capacidades de procesamiento multimodal optimizadas, adecuadas para una variedad de escenarios de tareas complejas."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash مدل اصلی پیشرفته گوگل است که به طور خاص برای استدلال پیشرفته، کدنویسی، ریاضیات و وظایف علمی طراحی شده است. این مدل دارای قابلیت «تفکر» داخلی است که به آن اجازه میدهد پاسخهایی با دقت بالاتر و پردازش زمینهای دقیقتری ارائه دهد.\n\nتوجه: این مدل دارای دو واریانت است: تفکر و غیرتفکر. قیمتگذاری خروجی بسته به فعال بودن قابلیت تفکر به طور قابل توجهی متفاوت است. اگر شما واریانت استاندارد (بدون پسوند «:thinking») را انتخاب کنید، مدل به وضوح از تولید توکنهای تفکر اجتناب خواهد کرد.\n\nبرای استفاده از قابلیت تفکر و دریافت توکنهای تفکر، شما باید واریانت «:thinking» را انتخاب کنید که منجر به قیمتگذاری بالاتر خروجی تفکر خواهد شد.\n\nعلاوه بر این، Gemini 2.5 Flash میتواند از طریق پارامتر «حداکثر تعداد توکنهای استدلال» پیکربندی شود، همانطور که در مستندات توضیح داده شده است (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview پیشرفتهترین مدل فکری گوگل است که قادر به استدلال درباره مسائل پیچیده در زمینه کد، ریاضیات و حوزههای STEM بوده و همچنین میتواند با استفاده از متنهای طولانی، مجموعههای داده بزرگ، کدها و مستندات را تحلیل کند."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash قابلیت پردازش چندوجهی بهینهشده را ارائه میدهد و برای انواع سناریوهای پیچیده مناسب است."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash est le modèle phare le plus avancé de Google, conçu pour des tâches de raisonnement avancé, de codage, de mathématiques et de sciences. Il comprend des capacités de 'pensée' intégrées, lui permettant de fournir des réponses avec une plus grande précision et un traitement contextuel détaillé.\n\nRemarque : ce modèle a deux variantes : pensée et non-pensée. La tarification de sortie varie considérablement en fonction de l'activation de la capacité de pensée. Si vous choisissez la variante standard (sans le suffixe ':thinking'), le modèle évitera explicitement de générer des jetons de pensée.\n\nPour tirer parti de la capacité de pensée et recevoir des jetons de pensée, vous devez choisir la variante ':thinking', ce qui entraînera une tarification de sortie de pensée plus élevée.\n\nDe plus, Gemini 2.5 Flash peut être configuré via le paramètre 'nombre maximal de jetons de raisonnement', comme décrit dans la documentation (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro est le modèle
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview est le modèle de pensée le plus avancé de Google, capable de raisonner sur des problèmes complexes en code, mathématiques et domaines STEM, ainsi que d'analyser de grands ensembles de données, des bases de code et des documents en utilisant un contexte étendu."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash propose des capacités de traitement multimodal optimisées, adaptées à divers scénarios de tâches complexes."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash è il modello principale più avanzato di Google, progettato per ragionamenti avanzati, codifica, matematica e compiti scientifici. Include capacità di 'pensiero' integrate, permettendo di fornire risposte con maggiore accuratezza e una gestione contestuale più dettagliata.\n\nNota: questo modello ha due varianti: pensiero e non pensiero. I prezzi di output variano significativamente a seconda che la capacità di pensiero sia attivata o meno. Se scegli la variante standard (senza il suffisso ':thinking'), il modello eviterà esplicitamente di generare token di pensiero.\n\nPer sfruttare la capacità di pensiero e ricevere token di pensiero, devi scegliere la variante ':thinking', che comporterà un prezzo di output di pensiero più elevato.\n\nInoltre, Gemini 2.5 Flash può essere configurato tramite il parametro 'numero massimo di token per il ragionamento', come descritto nella documentazione (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro è il modello
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview è il modello di pensiero più avanzato di Google, in grado di ragionare su problemi complessi nel campo del codice, della matematica e delle STEM, oltre a utilizzare un contesto esteso per analizzare grandi set di dati, repository di codice e documenti."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash offre capacità di elaborazione multimodale ottimizzate, adatte a vari scenari di compiti complessi."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flashは、Googleの最先端の主力モデルであり、高度な推論、コーディング、数学、科学タスクのために設計されています。内蔵の「思考」能力を備えており、より高い精度と詳細なコンテキスト処理で応答を提供します。\n\n注意：このモデルには、思考と非思考の2つのバリアントがあります。出力の価格は、思考能力が有効かどうかによって大きく異なります。標準バリアント（「:thinking」サフィックスなし）を選択すると、モデルは明示的に思考トークンの生成を避けます。\n\n思考能力を利用して思考トークンを受け取るには、「:thinking」バリアントを選択する必要があり、これにより思考出力の価格が高くなります。\n\nさらに、Gemini 2.5 Flashは、「推論最大トークン数」パラメータを介して構成可能であり、文書に記載されています (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)。"
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Proは、Google
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview は、Google の最先端の思考モデルであり、コード、数学、STEM 分野の複雑な問題を推論し、長いコンテキストを用いて大規模なデータセット、コードベース、ドキュメントを分析することができます。"
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flashは、最適化されたマルチモーダル処理能力を提供し、さまざまな複雑なタスクシナリオに適しています。"
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash는 Google의 최첨단 주력 모델로, 고급 추론, 코딩, 수학 및 과학 작업을 위해 설계되었습니다. 내장된 '사고' 능력을 포함하고 있어 더 높은 정확성과 세밀한 맥락 처리를 통해 응답을 제공합니다.\n\n주의: 이 모델에는 두 가지 변형이 있습니다: 사고 및 비사고. 출력 가격은 사고 능력이 활성화되었는지 여부에 따라 크게 다릅니다. 표준 변형(‘:thinking’ 접미사가 없는)을 선택하면 모델이 사고 토큰 생성을 명확히 피합니다.\n\n사고 능력을 활용하고 사고 토큰을 수신하려면 ‘:thinking’ 변형을 선택해야 하며, 이는 더 높은 사고 출력 가격을 발생시킵니다.\n\n또한, Gemini 2.5 Flash는 문서에 설명된 대로 '추론 최대 토큰 수' 매개변수를 통해 구성할 수 있습니다 (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro는 Google의 최첨단
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview는 Google의 최첨단 사고 모델로, 코드, 수학 및 STEM 분야의 복잡한 문제를 추론할 수 있으며, 긴 문맥을 사용하여 대규모 데이터 세트, 코드베이스 및 문서를 분석할 수 있습니다."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash는 최적화된 다중 모달 처리 능력을 제공하며, 다양한 복잡한 작업 시나리오에 적합합니다."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash is Google's meest geavanceerde hoofmodel, ontworpen voor geavanceerde redenering, codering, wiskunde en wetenschappelijke taken. Het bevat ingebouwde 'denkkracht', waardoor het in staat is om antwoorden te geven met een hogere nauwkeurigheid en gedetailleerde contextverwerking.\n\nLet op: dit model heeft twee varianten: denken en niet-denken. De outputprijs verschilt aanzienlijk afhankelijk van of de denkkracht is geactiveerd. Als u de standaardvariant kiest (zonder de ':thinking' suffix), zal het model expliciet vermijden om denk-tokens te genereren.\n\nOm gebruik te maken van de denkkracht en denk-tokens te ontvangen, moet u de ':thinking' variant kiezen, wat resulteert in hogere prijzen voor denk-output.\n\nBovendien kan Gemini 2.5 Flash worden geconfigureerd via de parameter 'max tokens for reasoning', zoals beschreven in de documentatie (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro is
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview is het meest geavanceerde denkkader van Google, dat in staat is om complexe problemen op het gebied van code, wiskunde en STEM te redeneren, en grote datasets, codebases en documenten te analyseren met behulp van lange context."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash biedt geoptimaliseerde multimodale verwerkingscapaciteiten, geschikt voor verschillende complexe taakscenario's."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "Gemini 2.5 Flash to najnowocześniejszy model główny Google, zaprojektowany z myślą o zaawansowanym wnioskowaniu, kodowaniu, matematyce i zadaniach naukowych. Zawiera wbudowaną zdolność 'myślenia', co pozwala mu na dostarczanie odpowiedzi z wyższą dokładnością i szczegółowym przetwarzaniem kontekstu.\n\nUwaga: ten model ma dwa warianty: myślenie i niemyslenie. Ceny wyjściowe różnią się znacznie w zależności od tego, czy zdolność myślenia jest aktywowana. Jeśli wybierzesz standardowy wariant (bez sufiksu ':thinking'), model wyraźnie unika generowania tokenów myślenia.\n\nAby skorzystać z zdolności myślenia i otrzymać tokeny myślenia, musisz wybrać wariant ':thinking', co spowoduje wyższe ceny wyjściowe za myślenie.\n\nPonadto Gemini 2.5 Flash można konfigurować za pomocą parametru 'maksymalna liczba tokenów do wnioskowania', jak opisano w dokumentacji (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "Gemini 2.5 Pro to najnowocześniejszy model
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview to najnowocześniejszy model myślowy Google, zdolny do rozumowania nad złożonymi problemami w dziedzinie kodowania, matematyki i STEM oraz do analizy dużych zbiorów danych, repozytoriów kodu i dokumentów przy użyciu długiego kontekstu."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash oferuje zoptymalizowane możliwości przetwarzania multimodalnego, odpowiednie do różnych złożonych scenariuszy zadań."
|
|
@@ -1187,8 +1187,8 @@
|
|
|
1187
1187
|
"google/gemini-2.5-flash-preview:thinking": {
|
|
1188
1188
|
"description": "O Gemini 2.5 Flash é o modelo principal mais avançado do Google, projetado para raciocínio avançado, codificação, matemática e tarefas científicas. Ele possui a capacidade de 'pensar' embutida, permitindo que forneça respostas com maior precisão e um tratamento de contexto mais detalhado.\n\nNota: Este modelo possui duas variantes: com e sem 'pensamento'. A precificação da saída varia significativamente dependendo da ativação da capacidade de pensamento. Se você escolher a variante padrão (sem o sufixo ':thinking'), o modelo evitará explicitamente gerar tokens de pensamento.\n\nPara aproveitar a capacidade de pensamento e receber tokens de pensamento, você deve escolher a variante ':thinking', que resultará em uma precificação de saída de pensamento mais alta.\n\nAlém disso, o Gemini 2.5 Flash pode ser configurado através do parâmetro 'número máximo de tokens para raciocínio', conforme descrito na documentação (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
|
|
1189
1189
|
},
|
|
1190
|
-
"google/gemini-2.5-pro-preview
|
|
1191
|
-
"description": "
|
|
1190
|
+
"google/gemini-2.5-pro-preview": {
|
|
1191
|
+
"description": "Gemini 2.5 Pro Preview é o modelo de pensamento mais avançado do Google, capaz de raciocinar sobre problemas complexos em código, matemática e áreas STEM, além de analisar grandes conjuntos de dados, bases de código e documentos usando contexto extenso."
|
|
1192
1192
|
},
|
|
1193
1193
|
"google/gemini-flash-1.5": {
|
|
1194
1194
|
"description": "Gemini 1.5 Flash oferece capacidades de processamento multimodal otimizadas, adequadas para uma variedade de cenários de tarefas complexas."
|