From 1b091df5291b22827aad9ee5dada9791ffc515ca Mon Sep 17 00:00:00 2001 From: wgqqqqq Date: Thu, 12 Mar 2026 20:35:48 +0800 Subject: [PATCH 1/2] Disable WebFetch by default in Cowork mode --- src/crates/core/src/agentic/agents/cowork_mode.rs | 1 - 1 file changed, 1 deletion(-) diff --git a/src/crates/core/src/agentic/agents/cowork_mode.rs b/src/crates/core/src/agentic/agents/cowork_mode.rs index 7edf4247..002d5de0 100644 --- a/src/crates/core/src/agentic/agents/cowork_mode.rs +++ b/src/crates/core/src/agentic/agents/cowork_mode.rs @@ -32,7 +32,6 @@ impl CoworkMode { "Git".to_string(), "Bash".to_string(), "TerminalControl".to_string(), - "WebFetch".to_string(), "WebSearch".to_string(), ], } From 0cba66ebe688308eedd117c66d639462fe868392 Mon Sep 17 00:00:00 2001 From: wgqqqqq Date: Thu, 12 Mar 2026 20:37:26 +0800 Subject: [PATCH 2/2] Update web tool defaults and model config settings --- .../tools/implementations/web_tools.rs | 493 ++++++++---------- .../config/components/AIModelConfig.tsx | 18 +- .../config/components/DefaultModelConfig.tsx | 4 - .../src/infrastructure/config/types/index.ts | 10 - src/web-ui/src/locales/en-US/settings.json | 3 - .../src/locales/en-US/settings/ai-model.json | 5 - .../locales/en-US/settings/default-model.json | 4 - src/web-ui/src/locales/zh-CN/settings.json | 3 - .../src/locales/zh-CN/settings/ai-model.json | 5 - .../locales/zh-CN/settings/default-model.json | 4 - 10 files changed, 232 insertions(+), 317 deletions(-) diff --git a/src/crates/core/src/agentic/tools/implementations/web_tools.rs b/src/crates/core/src/agentic/tools/implementations/web_tools.rs index 0de21b2b..6f2f546d 100644 --- a/src/crates/core/src/agentic/tools/implementations/web_tools.rs +++ b/src/crates/core/src/agentic/tools/implementations/web_tools.rs @@ -1,40 +1,34 @@ //! Web tool implementation - WebSearchTool and URLFetcherTool use crate::agentic::tools::framework::{Tool, ToolResult, ToolUseContext, ValidationResult}; -use crate::infrastructure::get_path_manager_arc; -use crate::service::config::types::GlobalConfig; use crate::util::errors::{BitFunError, BitFunResult}; use async_trait::async_trait; -use log::{debug, error, info, warn}; +use log::{error, info}; use serde::Deserialize; use serde_json::{json, Value}; -use std::env; -use std::fs; +use std::time::Duration; + +const EXA_URL: &str = "https://mcp.exa.ai/mcp"; +const EXA_RESULTS: u64 = 5; +const EXA_CONTEXT: u64 = 8_000; -/// ZhipuAI Web Search API response #[derive(Debug, Deserialize)] -struct ZhipuSearchResponse { - search_result: Vec, +struct ExaRes { + result: Option, } #[derive(Debug, Deserialize)] -struct ZhipuSearchResult { - title: String, - content: String, - link: String, - #[serde(default)] - media: String, +struct ExaData { + content: Vec, } -/// Search API configuration -#[derive(Debug, Clone)] -struct SearchApiConfig { - api_key: String, - base_url: String, - model_name: String, +#[derive(Debug, Deserialize)] +struct ExaContent { + #[serde(rename = "type")] + kind: String, + text: Option, } -/// Web search tool - supports reading API configuration from config file or environment variables pub struct WebSearchTool; impl WebSearchTool { @@ -42,191 +36,166 @@ impl WebSearchTool { Self } - /// Load search API configuration from config file - async fn load_search_config(&self) -> Option { - // 1. Prefer environment variables - if let Ok(api_key) = env::var("ZHIPU_API_KEY").or_else(|_| env::var("API_KEY")) { - info!("WebSearchTool: Loaded API key from environment variable"); - return Some(SearchApiConfig { - api_key, - base_url: "https://open.bigmodel.cn/api/paas/v4/web_search".to_string(), - model_name: "ZhipuAI Web Search".to_string(), - }); - } - - // 2. Read from config file - match self.load_config_from_file().await { - Ok(Some(config)) => { - info!( - "WebSearchTool: Loaded search API config from file: {}", - config.model_name - ); - Some(config) - } - Ok(None) => { - warn!("WebSearchTool: Search API not configured, will return mock data"); - None - } - Err(e) => { - warn!( - "WebSearchTool: Failed to load config, will return mock data: {}", - e - ); - None - } - } - } - - /// Load from config file - async fn load_config_from_file(&self) -> BitFunResult> { - // Get config file path - let path_manager = get_path_manager_arc(); - let config_file = path_manager.app_config_file(); - - if !config_file.exists() { - debug!("Config file does not exist: {:?}", config_file); - return Ok(None); - } - - // Read and parse config file - let config_content = fs::read_to_string(&config_file) - .map_err(|e| BitFunError::tool(format!("Failed to read config file: {}", e)))?; - - let global_config: GlobalConfig = serde_json::from_str(&config_content) - .map_err(|e| BitFunError::tool(format!("Failed to parse config file: {}", e)))?; - - // Get search model ID - let search_model_id = match global_config.ai.default_models.search { - Some(id) if !id.is_empty() => id, - _ => { - debug!("Search model not configured"); - return Ok(None); - } - }; - - // Find corresponding model configuration - let model_config = global_config - .ai - .models - .iter() - .find(|m| m.id == search_model_id) - .ok_or_else(|| { - BitFunError::tool(format!( - "Search model config not found: {}", - search_model_id - )) - })?; - - // Validate API Key - if model_config.api_key.trim().is_empty() { - warn!("Search model API key not configured: {}", model_config.name); - return Ok(None); - } - - Ok(Some(SearchApiConfig { - api_key: model_config.api_key.clone(), - base_url: model_config.base_url.clone(), - model_name: model_config.name.clone(), - })) - } - - /// Call search API (supports ZhipuAI and other compatible search APIs) - async fn search_api( + async fn search( &self, - config: &SearchApiConfig, query: &str, - count: u64, - search_engine: &str, - content_size: &str, - recency_filter: &str, - ) -> BitFunResult> { - info!( - "Search API call: model={}, query={}, engine={}, content_size={}, recency={}", - config.model_name, query, search_engine, content_size, recency_filter - ); - - // Create HTTP client - let client = reqwest::Client::builder() - .timeout(std::time::Duration::from_secs(30)) + num: u64, + kind: &str, + crawl: &str, + ctx: u64, + ) -> BitFunResult { + let cli = reqwest::Client::builder() + .timeout(Duration::from_secs(25)) .build() - .map_err(|e| BitFunError::tool(format!("Failed to create HTTP client: {}", e)))?; - - // Build request (ZhipuAI format) - let request_body = json!({ - "search_query": query, - "search_engine": search_engine, - "search_intent": true, // Always enable search intent recognition - "count": count, - "content_size": content_size, - "search_recency_filter": recency_filter + .map_err(|err| BitFunError::tool(format!("Failed to create HTTP client: {}", err)))?; + + let body = json!({ + "jsonrpc": "2.0", + "id": 1, + "method": "tools/call", + "params": { + "name": "web_search_exa", + "arguments": { + "query": query, + "type": kind, + "numResults": num, + "livecrawl": crawl, + "contextMaxCharacters": ctx, + } + } }); - debug!( - "Request body: {}", - serde_json::to_string_pretty(&request_body).unwrap_or_default() - ); - - // Send request - let response = client - .post(&config.base_url) - .header("Authorization", format!("Bearer {}", config.api_key)) - .header("Content-Type", "application/json") - .json(&request_body) + let res = cli + .post(EXA_URL) + .header("accept", "application/json, text/event-stream") + .header("content-type", "application/json") + .json(&body) .send() .await - .map_err(|e| BitFunError::tool(format!("Failed to send request: {}", e)))?; + .map_err(|err| BitFunError::tool(format!("Failed to send request: {}", err)))?; - // Check status code - let status = response.status(); + let status = res.status(); if !status.is_success() { - let error_text = response + let err = res .text() .await .unwrap_or_else(|_| String::from("Unknown error")); - error!("Search API error: status={}, error={}", status, error_text); + error!("WebSearch Exa error: status={}, error={}", status, err); return Err(BitFunError::tool(format!( - "Search API error {}: {}", - status, error_text + "Web search error {}: {}", + status, err ))); } - // Parse response - let search_response: ZhipuSearchResponse = response - .json() + let text = res + .text() .await - .map_err(|e| BitFunError::tool(format!("Failed to parse response: {}", e)))?; + .map_err(|err| BitFunError::tool(format!("Failed to read response: {}", err)))?; - info!( - "Found {} results from {}", - search_response.search_result.len(), - config.model_name - ); + self.parse_sse(&text) + } - // Convert to unified format - let results: Vec = search_response - .search_result - .into_iter() - .map(|r| { - json!({ - "title": r.title, - "url": r.link, - "snippet": r.content, - "source": r.media - }) - }) - .collect(); + fn parse_sse(&self, text: &str) -> BitFunResult { + let out = text + .lines() + .filter_map(|line| line.strip_prefix("data: ")) + .find_map(|line| { + serde_json::from_str::(line) + .ok() + .and_then(|res| res.result) + .map(|res| { + res.content + .into_iter() + .filter(|item| item.kind == "text") + .filter_map(|item| item.text) + .collect::>() + .join("\n") + }) + .filter(|item| !item.trim().is_empty()) + }); - Ok(results) + out.ok_or_else(|| BitFunError::tool("Web search returned no content".to_string())) } - /// Return mock search results (when search API is not configured) - fn mock_search(&self, query: &str) -> Vec { - vec![json!({ - "title": format!("Search results for: {}", query), - "url": "https://example.com", - "snippet": "This is mock data. To get real search results, please configure as follows:\n\n1. Open Config Center → AI Model Configuration\n2. Create new configuration, select \"Search Enhancement Model\" category\n3. Fill in Search API URL and API Key (supports ZhipuAI, etc.)\n4. In Config Center → Super Agent, select the search model you just configured\n\nOr set environment variable ZHIPU_API_KEY (visit https://open.bigmodel.cn/ to get)", - "source": "Mock Data" - })] + fn results(&self, text: &str) -> Vec { + let mut out = Vec::new(); + let mut cur: Option<(String, String, Vec)> = None; + let mut body = false; + + for line in text.lines() { + if let Some(next) = line.strip_prefix("Title: ") { + if let Some((title, url, text)) = cur.take() { + out.push(self.item(title, url, text)); + } + cur = Some((next.trim().to_string(), String::new(), Vec::new())); + body = false; + continue; + } + + let Some(cur) = cur.as_mut() else { + continue; + }; + + if let Some(next) = line.strip_prefix("URL: ") { + cur.1 = next.trim().to_string(); + continue; + } + + if let Some(next) = line.strip_prefix("Text: ") { + if !next.trim().is_empty() { + cur.2.push(next.trim().to_string()); + } + body = true; + continue; + } + + if body { + cur.2.push(line.to_string()); + } + } + + if let Some((title, url, text)) = cur.take() { + out.push(self.item(title, url, text)); + } + + if out.is_empty() && !text.trim().is_empty() { + return vec![json!({ + "title": "Web search result", + "url": "", + "snippet": self.snippet(text) + })]; + } + + out + } + + fn item(&self, title: String, url: String, text: Vec) -> Value { + json!({ + "title": title, + "url": url, + "snippet": self.snippet(&text.join("\n")) + }) + } + + fn snippet(&self, text: &str) -> String { + let text = text + .lines() + .map(str::trim) + .filter(|line| !line.is_empty()) + .filter(|line| !line.starts_with('#')) + .collect::>() + .join(" "); + + if text.chars().count() <= 320 { + return text; + } + + let mut out = String::new(); + for ch in text.chars().take(317) { + out.push(ch); + } + out.push_str("..."); + out } } @@ -240,6 +209,7 @@ impl Tool for WebSearchTool { Ok( r#"- Allows BitFun to search the web and use the results to inform responses - Provides up-to-date information for current events and recent data +- Uses Exa's hosted MCP web search service with no local API key setup - Returns search result information formatted as search result blocks - Use this tool for accessing information beyond BitFun's knowledge cutoff @@ -250,11 +220,10 @@ Usage notes: - Results include title, URL, snippet and source information Advanced features: -- Automatically uses advanced search engine (search_pro) for best results -- Control content size: small (brief), medium (moderate), high (detailed) -- Filter by recency: oneDay, oneWeek, oneMonth, oneYear, or noLimit -- Return up to 50 results per query -- Search intent recognition is automatically enabled"# +- Choose search depth: auto, fast, or deep +- Control result count and context size for LLM-friendly output +- Optionally prefer live crawling for fresher pages +- Return up to 10 results per query"# .to_string(), ) } @@ -269,28 +238,29 @@ Advanced features: }, "num_results": { "type": "number", - "description": "Number of search results to return (1-50, default: 5)", - "default": 5, + "description": "Number of search results to return (1-10, default: 5)", + "default": EXA_RESULTS, "minimum": 1, - "maximum": 50 + "maximum": 10 }, - "search_engine": { + "type": { "type": "string", - "enum": ["search_std", "search_pro", "search_pro_sogou", "search_pro_quark"], - "description": "Search engine to use. MUST be one of: 'search_std' (standard engine), 'search_pro' (default, advanced engine), 'search_pro_sogou' (Sogou search), 'search_pro_quark' (Quark search). No other values are accepted.", - "default": "search_pro" + "enum": ["auto", "fast", "deep"], + "description": "Search depth. Use 'auto' for balanced results, 'fast' for lower latency, or 'deep' for broader context.", + "default": "auto" }, - "content_size": { + "livecrawl": { "type": "string", - "enum": ["small", "medium", "high"], - "description": "Content snippet size. MUST be one of: 'small' (brief snippets), 'medium' (default, moderate length), 'high' (detailed content). No other values are accepted.", - "default": "medium" + "enum": ["fallback", "preferred"], + "description": "Live crawl mode. Use 'preferred' to favor fresh crawling, or 'fallback' to use cached data when possible.", + "default": "fallback" }, - "search_recency_filter": { - "type": "string", - "enum": ["noLimit", "oneDay", "oneWeek", "oneMonth", "oneYear"], - "description": "Filter results by time range. MUST be one of: 'noLimit' (default, no time limit), 'oneDay' (past 24 hours), 'oneWeek' (past 7 days), 'oneMonth' (past 30 days), 'oneYear' (past 365 days). No other values are accepted.", - "default": "noLimit" + "context_max_characters": { + "type": "number", + "description": "Maximum characters of search context to request from Exa (default: 8000)", + "default": EXA_CONTEXT, + "minimum": 1000, + "maximum": 20000 } }, "required": ["query"] @@ -322,63 +292,30 @@ Advanced features: let num_results = input .get("num_results") .and_then(|v| v.as_u64()) - .unwrap_or(5) - .min(50) // Limit to maximum 50 results - .max(1); // At least 1 result + .unwrap_or(EXA_RESULTS) + .clamp(1, 10); - // Get optional parameters - let search_engine = input - .get("search_engine") - .and_then(|v| v.as_str()) - .unwrap_or("search_pro"); + let kind = input.get("type").and_then(|v| v.as_str()).unwrap_or("auto"); - let content_size = input - .get("content_size") + let crawl = input + .get("livecrawl") .and_then(|v| v.as_str()) - .unwrap_or("medium"); + .unwrap_or("fallback"); - let recency_filter = input - .get("search_recency_filter") - .and_then(|v| v.as_str()) - .unwrap_or("noLimit"); + let ctx = input + .get("context_max_characters") + .and_then(|v| v.as_u64()) + .unwrap_or(EXA_CONTEXT) + .clamp(1_000, 20_000); info!( - "WebSearch tool called: query='{}', num_results={}", - query, num_results + "WebSearch Exa call: query='{}', num_results={}, type={}, livecrawl={}, context_max_characters={}", + query, num_results, kind, crawl, ctx ); - // Load search API configuration - let search_config = self.load_search_config().await; - - // Try real search, return mock data on failure - let (results, api_used) = if let Some(config) = search_config { - info!("Using search API: {}", config.model_name); - match self - .search_api( - &config, - query, - num_results, - search_engine, - content_size, - recency_filter, - ) - .await - { - Ok(results) => { - info!("Search succeeded, returning {} results", results.len()); - (results, true) - } - Err(e) => { - warn!("Search failed, returning mock data: {}", e); - (self.mock_search(query), false) - } - } - } else { - info!("Search API not configured, returning mock data"); - (self.mock_search(query), false) - }; + let raw = self.search(query, num_results, kind, crawl, ctx).await?; + let results = self.results(&raw); - // Format results let formatted_results = results .iter() .enumerate() @@ -399,7 +336,7 @@ Advanced features: "query": query, "results": results, "result_count": results.len(), - "api_used": api_used + "provider": "exa_mcp" }), result_for_assistant: Some(format!( "Search query: '{}'\nFound {} results:\n\n{}", @@ -601,7 +538,7 @@ Example usage: #[cfg(test)] mod tests { - use super::WebFetchTool; + use super::{WebFetchTool, WebSearchTool}; use crate::agentic::tools::framework::{Tool, ToolResult, ToolUseContext}; use serde_json::json; use std::collections::HashMap; @@ -667,9 +604,12 @@ mod tests { "format": "text" }); - let results = tool.call(&input, &empty_context()).await.unwrap_or_else(|e| { - panic!("tool call failed with detailed error: {:?}", e); - }); + let results = tool + .call(&input, &empty_context()) + .await + .unwrap_or_else(|e| { + panic!("tool call failed with detailed error: {:?}", e); + }); assert_eq!(results.len(), 1); match &results[0] { @@ -679,10 +619,7 @@ mod tests { } => { assert_eq!(data["content"], "hello from webfetch"); assert_eq!(data["format"], "text"); - assert_eq!( - result_for_assistant.as_deref(), - Some("hello from webfetch") - ); + assert_eq!(result_for_assistant.as_deref(), Some("hello from webfetch")); } other => panic!("unexpected tool result variant: {:?}", other), } @@ -699,9 +636,12 @@ mod tests { "format": "text" }); - let results = tool.call(&input, &empty_context()).await.unwrap_or_else(|e| { - panic!("tool call failed with detailed error: {:?}", e); - }); + let results = tool + .call(&input, &empty_context()) + .await + .unwrap_or_else(|e| { + panic!("tool call failed with detailed error: {:?}", e); + }); assert_eq!(results.len(), 1); match &results[0] { @@ -722,4 +662,27 @@ mod tests { } } + #[test] + fn websearch_parses_exa_text_into_results() { + let tool = WebSearchTool::new(); + let text = r#"Title: Result One +URL: https://example.com/one +Text: Result One + +First paragraph. + +Title: Result Two +URL: https://example.com/two +Text: Result Two + +Second paragraph. +"#; + + let out = tool.results(text); + assert_eq!(out.len(), 2); + assert_eq!(out[0]["title"], "Result One"); + assert_eq!(out[0]["url"], "https://example.com/one"); + assert_eq!(out[0]["snippet"], "Result One First paragraph."); + assert_eq!(out[1]["title"], "Result Two"); + } } diff --git a/src/web-ui/src/infrastructure/config/components/AIModelConfig.tsx b/src/web-ui/src/infrastructure/config/components/AIModelConfig.tsx index 63285157..e80f0f76 100644 --- a/src/web-ui/src/infrastructure/config/components/AIModelConfig.tsx +++ b/src/web-ui/src/infrastructure/config/components/AIModelConfig.tsx @@ -248,7 +248,7 @@ const AIModelConfig: React.FC = () => { } - if (editingConfig.category !== 'search_enhanced' && !editingConfig.model_name) { + if (!editingConfig.model_name) { notification.warning(t('messages.fillModelName')); return; } @@ -598,15 +598,6 @@ const AIModelConfig: React.FC = () => { updates.base_url = 'https://open.bigmodel.cn/api/paas/v4/images/generations'; updates.request_url = resolveRequestUrl(updates.base_url!, prev?.provider || 'openai', prev?.model_name || ''); break; - case 'search_enhanced': - defaultCapabilities = ['search']; - updates.base_url = 'https://open.bigmodel.cn/api/paas/v4/web_search'; - updates.request_url = resolveRequestUrl(updates.base_url!, 'openai'); - updates.model_name = 'search-api'; - updates.provider = 'openai'; - updates.context_window = 128000; - updates.max_tokens = 8192; - break; case 'speech_recognition': defaultCapabilities = ['speech_recognition']; updates.base_url = 'https://open.bigmodel.cn/api/paas/v4/chat/completions'; @@ -741,14 +732,13 @@ const AIModelConfig: React.FC = () => { { label: t('category.general_chat'), value: 'general_chat' }, { label: t('category.multimodal'), value: 'multimodal' }, { label: t('category.image_generation'), value: 'image_generation' }, - { label: t('category.search_enhanced'), value: 'search_enhanced' }, { label: t('category.speech_recognition'), value: 'speech_recognition' }, ]} /> setEditingConfig(prev => ({ ...prev, name: e.target.value }))} placeholder={t('form.configNamePlaceholder')} inputSize="small" /> - +
{ request_url: resolveRequestUrl(e.target.value, prev?.provider || 'openai', prev?.model_name || '') }))} onFocus={(e) => e.target.select()} - placeholder={editingConfig.category === 'search_enhanced' ? 'https://open.bigmodel.cn/api/paas/v4/web_search' : 'https://open.bigmodel.cn/api/paas/v4/chat/completions'} + placeholder={'https://open.bigmodel.cn/api/paas/v4/chat/completions'} inputSize="small" /> {editingConfig.base_url && ( @@ -779,7 +769,7 @@ const AIModelConfig: React.FC = () => { )} - {!isFromTemplate && editingConfig.category !== 'search_enhanced' && ( + {!isFromTemplate && ( <> setEditingConfig(prev => ({ ...prev, model_name: e.target.value, request_url: resolveRequestUrl(prev?.base_url || '', prev?.provider || 'openai', e.target.value) }))} placeholder={editingConfig.category === 'speech_recognition' ? 'glm-asr' : 'glm-4.7'} inputSize="small" /> diff --git a/src/web-ui/src/infrastructure/config/components/DefaultModelConfig.tsx b/src/web-ui/src/infrastructure/config/components/DefaultModelConfig.tsx index abacec73..32c69ebd 100644 --- a/src/web-ui/src/infrastructure/config/components/DefaultModelConfig.tsx +++ b/src/web-ui/src/infrastructure/config/components/DefaultModelConfig.tsx @@ -24,7 +24,6 @@ const log = createLogger('DefaultModelConfig'); const OPTIONAL_CAPABILITY_TYPES: OptionalCapabilityType[] = [ 'image_understanding', 'image_generation', - 'search', 'speech_recognition' ]; @@ -73,7 +72,6 @@ export const DefaultModelConfig: React.FC = () => { setOptionalCapabilities({ image_understanding: defaultModelsConfig?.image_understanding, image_generation: defaultModelsConfig?.image_generation, - search: defaultModelsConfig?.search, speech_recognition: defaultModelsConfig?.speech_recognition, }); } catch (error) { @@ -156,8 +154,6 @@ export const DefaultModelConfig: React.FC = () => { return m.capabilities?.includes('image_understanding') || m.category === 'multimodal'; case 'image_generation': return m.capabilities?.includes('image_generation') || m.category === 'image_generation'; - case 'search': - return m.capabilities?.includes('search') || m.category === 'search_enhanced'; case 'speech_recognition': return m.capabilities?.includes('speech_recognition') || m.category === 'speech_recognition'; default: diff --git a/src/web-ui/src/infrastructure/config/types/index.ts b/src/web-ui/src/infrastructure/config/types/index.ts index c53cd8ab..658688e7 100644 --- a/src/web-ui/src/infrastructure/config/types/index.ts +++ b/src/web-ui/src/infrastructure/config/types/index.ts @@ -70,7 +70,6 @@ export type ModelCapability = | 'text_chat' | 'image_understanding' | 'image_generation' - | 'search' | 'function_calling' | 'speech_recognition'; @@ -78,7 +77,6 @@ export type ModelCategory = | 'general_chat' | 'multimodal' | 'image_generation' - | 'search_enhanced' | 'speech_recognition'; export interface ModelMetadata { @@ -93,7 +91,6 @@ export const CAPABILITY_LABELS: Record = { text_chat: t('settings/ai-model:capabilities.text_chat'), image_understanding: t('settings/ai-model:capabilities.image_understanding'), image_generation: t('settings/ai-model:capabilities.image_generation'), - search: t('settings/ai-model:capabilities.search'), function_calling: t('settings/ai-model:capabilities.function_calling'), speech_recognition: t('settings/ai-model:capabilities.speech_recognition') }; @@ -103,7 +100,6 @@ export const CATEGORY_LABELS: Record = { general_chat: t('settings/ai-model:category.general_chat'), multimodal: t('settings/ai-model:category.multimodal'), image_generation: t('settings/ai-model:category.image_generation'), - search_enhanced: t('settings/ai-model:category.search_enhanced'), speech_recognition: t('settings/ai-model:category.speech_recognition') }; @@ -112,7 +108,6 @@ export const CATEGORY_ICONS: Record = { general_chat: t('settings/ai-model:categoryIcons.general_chat'), multimodal: t('settings/ai-model:categoryIcons.multimodal'), image_generation: t('settings/ai-model:categoryIcons.image_generation'), - search_enhanced: t('settings/ai-model:categoryIcons.search_enhanced'), speech_recognition: t('settings/ai-model:categoryIcons.speech_recognition') }; @@ -174,8 +169,6 @@ export interface DefaultModelsConfig { fast?: string | null; - search?: string | null; - image_understanding?: string | null; speech_recognition?: string | null; @@ -594,8 +587,6 @@ export interface OptionalCapabilityModels { image_generation?: string; - search?: string; - speech_recognition?: string; } @@ -603,5 +594,4 @@ export interface OptionalCapabilityModels { export type OptionalCapabilityType = | 'image_understanding' | 'image_generation' - | 'search' | 'speech_recognition'; diff --git a/src/web-ui/src/locales/en-US/settings.json b/src/web-ui/src/locales/en-US/settings.json index 4b8f3538..18b6a8d7 100644 --- a/src/web-ui/src/locales/en-US/settings.json +++ b/src/web-ui/src/locales/en-US/settings.json @@ -240,18 +240,15 @@ "general_chat": "Text Generation", "multimodal": "Image Understanding", "image_generation": "Image Generation", - "search_enhanced": "Information Retrieval", "speech_recognition": "Speech Recognition" }, "categoryHints": { "general_chat": "Text Generation: Generate text responses, code, etc. for most conversation scenarios", "multimodal": "Image Understanding: Understand image content and have mixed text-image conversations", "image_generation": "Image Generation: Generate images from text descriptions", - "search_enhanced": "Information Retrieval: Search the web for real-time information, only need name, API URL and key", "speech_recognition": "Speech Recognition: Convert speech to text (e.g., GLM-ASR)" }, "modelNameHint": "GLM-ASR uses glm-asr, other speech models use their corresponding model names", - "searchApiHint": "Zhipu AI search API URL, or other compatible search API URL", "thinking": { "enable": "Enable Thinking", "enableHint": "Requires model to support thinking mode", diff --git a/src/web-ui/src/locales/en-US/settings/ai-model.json b/src/web-ui/src/locales/en-US/settings/ai-model.json index af248a76..8119eb14 100644 --- a/src/web-ui/src/locales/en-US/settings/ai-model.json +++ b/src/web-ui/src/locales/en-US/settings/ai-model.json @@ -76,21 +76,18 @@ "general_chat": "Text Generation", "multimodal": "Image Understanding", "image_generation": "Image Generation", - "search_enhanced": "Information Retrieval", "speech_recognition": "Speech Recognition" }, "categoryIcons": { "general_chat": "Text", "multimodal": "Vision", "image_generation": "Image", - "search_enhanced": "Search", "speech_recognition": "Speech" }, "categoryHints": { "general_chat": "Text Generation: Generate text responses, code, etc. for most conversation scenarios", "multimodal": "Image Understanding: Understand image content and have mixed text-image conversations", "image_generation": "Image Generation: Generate images from text descriptions", - "search_enhanced": "Information Retrieval: Search the web for real-time information, only need name, API URL and key", "speech_recognition": "Speech Recognition: Convert speech to text (e.g., GLM-ASR)" }, "form": { @@ -101,7 +98,6 @@ "baseUrl": "API URL", "resolvedUrlLabel": "Request URL: ", "forceUrlHint": "Append # to force use the exact URL you entered", - "searchApiHint": "Zhipu AI search API URL, or other compatible search API URL", "apiKey": "API Key", "apiKeyPlaceholder": "Enter your API Key", "provider": "Request Format", @@ -152,7 +148,6 @@ "text_chat": "Chat", "image_understanding": "Vision", "image_generation": "Image", - "search": "Search", "function_calling": "Tools", "speech_recognition": "Speech" }, diff --git a/src/web-ui/src/locales/en-US/settings/default-model.json b/src/web-ui/src/locales/en-US/settings/default-model.json index 6806a9f4..cc8d4ab4 100644 --- a/src/web-ui/src/locales/en-US/settings/default-model.json +++ b/src/web-ui/src/locales/en-US/settings/default-model.json @@ -35,10 +35,6 @@ "label": "Image Generation", "description": "Generate images from text descriptions" }, - "search": { - "label": "Search Enhanced", - "description": "Search the web in real-time for latest information" - }, "speech_recognition": { "label": "Speech Recognition", "description": "Convert speech to text, support voice input" diff --git a/src/web-ui/src/locales/zh-CN/settings.json b/src/web-ui/src/locales/zh-CN/settings.json index f73c43fc..d8e6eeea 100644 --- a/src/web-ui/src/locales/zh-CN/settings.json +++ b/src/web-ui/src/locales/zh-CN/settings.json @@ -240,18 +240,15 @@ "general_chat": "文本生成", "multimodal": "图像理解", "image_generation": "图像生成", - "search_enhanced": "信息检索", "speech_recognition": "语音识别" }, "categoryHints": { "general_chat": "文本生成：生成文本回复、代码等，适用于大多数对话场景", "multimodal": "图像理解：理解图片内容并进行图文混合对话", "image_generation": "图像生成：根据文字描述生成图片", - "search_enhanced": "信息检索：搜索网络获取实时信息，只需配置名称、API地址和密钥", "speech_recognition": "语音识别：将语音转换为文字（如智谱 GLM-ASR）" }, "modelNameHint": "智谱 GLM-ASR 使用 glm-asr，其他语音模型填写对应的模型名称", - "searchApiHint": "智谱AI搜索API地址，或其他兼容的搜索API地址", "thinking": { "enable": "启用思考", "enableHint": "需要模型支持思考模式", diff --git a/src/web-ui/src/locales/zh-CN/settings/ai-model.json b/src/web-ui/src/locales/zh-CN/settings/ai-model.json index c8e6e96e..22000920 100644 --- a/src/web-ui/src/locales/zh-CN/settings/ai-model.json +++ b/src/web-ui/src/locales/zh-CN/settings/ai-model.json @@ -76,21 +76,18 @@ "general_chat": "文本生成", "multimodal": "多模态", "image_generation": "图像生成", - "search_enhanced": "信息检索", "speech_recognition": "语音识别" }, "categoryIcons": { "general_chat": "文本", "multimodal": "多模态", "image_generation": "绘图", - "search_enhanced": "检索", "speech_recognition": "语音" }, "categoryHints": { "general_chat": "文本生成：生成文本回复、代码等，适用于大多数对话场景", "multimodal": "多模态：理解图片内容并进行图文混合对话", "image_generation": "图像生成：根据文字描述生成图片", - "search_enhanced": "信息检索：搜索网络获取实时信息，只需配置名称、API地址和密钥", "speech_recognition": "语音识别：将语音转换为文字（如智谱 GLM-ASR）" }, "form": { @@ -101,7 +98,6 @@ "baseUrl": "API地址", "resolvedUrlLabel": "实际请求地址：", "forceUrlHint": "以 # 结尾可强制使用输入地址", - "searchApiHint": "智谱AI搜索API地址，或其他兼容的搜索API地址", "apiKey": "API密钥", "apiKeyPlaceholder": "输入您的 API Key", "provider": "请求格式", @@ -152,7 +148,6 @@ "text_chat": "对话", "image_understanding": "多模态", "image_generation": "绘图", - "search": "搜索", "function_calling": "工具", "speech_recognition": "语音" }, diff --git a/src/web-ui/src/locales/zh-CN/settings/default-model.json b/src/web-ui/src/locales/zh-CN/settings/default-model.json index 6ae7d1f4..df87c59f 100644 --- a/src/web-ui/src/locales/zh-CN/settings/default-model.json +++ b/src/web-ui/src/locales/zh-CN/settings/default-model.json @@ -35,10 +35,6 @@ "label": "图像生成", "description": "根据文字描述生成图片" }, - "search": { - "label": "搜索增强", - "description": "实时搜索网络信息，获取最新数据" - }, "speech_recognition": { "label": "语音识别", "description": "将语音转换为文字，支持语音输入"