diff --git a/README.md b/README.md
index 150f09f..69fe154 100644
--- a/README.md
+++ b/README.md
@@ -300,6 +300,7 @@ python scripts/apply_term_suggestions.py ^
 - `ARTICLE_SUMMARY_LLM_API_KEY`
 
 如果这些变量未设置，单篇总结会回退使用主流程中的 `LLM_*` / `OPENAI_*` 配置。
+如果显式指定 DeepSeek 作为单篇总结模型且请求超时或失败，当前实现会自动再用主流程默认模型配置重试一次。
 
 示例（PowerShell 风格）：
 
@@ -316,7 +317,8 @@ set ARTICLE_SUMMARY_LLM_MODEL=deepseek-chat
 python scripts/run_article_summaries.py ^
   --extracted outputs/freshrss/extracted/freshrss.extracted.json ^
   --ids 12345 67890 ^
-  --output-dir outputs/freshrss/single_summaries
+  --output-dir outputs/freshrss/single_summaries ^
+  --timeout 120
 ```
 
 也可以通过 `summary_mcp.server` 暴露的 MCP 工具 `generate_article_summaries` 调用：
diff --git a/scripts/run_article_summaries.py b/scripts/run_article_summaries.py
index ccd7423..b014824 100755
--- a/scripts/run_article_summaries.py
+++ b/scripts/run_article_summaries.py
@@ -37,7 +37,7 @@ def main() -> None:
         help="Directory to write per-article Markdown summaries",
     )
     parser.add_argument("--max-retries", type=int, default=2, help="Maximum LLM retry attempts per article")
-    parser.add_argument("--timeout", type=float, default=60.0, help="LLM request timeout in seconds")
+    parser.add_argument("--timeout", type=float, default=120.0, help="LLM request timeout in seconds")
     parser.add_argument("--api-key", type=str, default=None, help="Override article-summary LLM API key")
     parser.add_argument("--model", type=str, default=None, help="Override article-summary LLM model")
     parser.add_argument("--api-url", type=str, default=None, help="Override article-summary LLM API URL/base URL")
diff --git a/src/summary_mcp/server.py b/src/summary_mcp/server.py
index 05c2f16..6e02a6f 100644
--- a/src/summary_mcp/server.py
+++ b/src/summary_mcp/server.py
@@ -172,7 +172,7 @@ def generate_article_summaries(
     selected_ids: list[str],
     output_dir: str | None = None,
     max_retries: int = 2,
-    timeout_seconds: float = 60.0,
+    timeout_seconds: float = 120.0,
     llm_api_key: str | None = None,
     llm_model: str | None = None,
     llm_api_url: str | None = None,
diff --git a/src/summary_mcp/workflows/article_summary.py b/src/summary_mcp/workflows/article_summary.py
index 0183d95..98214d5 100644
--- a/src/summary_mcp/workflows/article_summary.py
+++ b/src/summary_mcp/workflows/article_summary.py
@@ -4,6 +4,8 @@ from dataclasses import dataclass
 from pathlib import Path
 from typing import Iterable, Mapping, Sequence
 
+import httpx
+
 from summary_mcp.core.summary_loop import run_loop_payload
 from summary_mcp.validators.article_summary import validate_article_summary_payload
 
@@ -18,7 +20,7 @@ DEFAULT_PROMPT_PATH = OUTPUT_ROOT / "prompts" / "article-summary-prompt.txt"
 class ArticleSummaryConfig:
     prompt_path: Path = DEFAULT_PROMPT_PATH
     max_retries: int = 2
-    timeout_seconds: float = 60.0
+    timeout_seconds: float = 120.0
     llm_api_key: str | None = None
     llm_model: str | None = None
     llm_api_url: str | None = None
@@ -187,17 +189,45 @@ def summarize_selected_articles(
             "warnings": warnings,
         }
 
-        summary_exit_code, summary_payload, _ = run_loop_payload(
-            extracted_payload=extracted_payload,
-            prompt_path=cfg.prompt_path,
-            max_retries=cfg.max_retries,
-            timeout_seconds=cfg.timeout_seconds,
-            api_key=resolved_api_key,
-            model=resolved_model,
-            api_url=resolved_api_url,
-            output_path=None,
-            validator=validate_article_summary_payload,
+        summary_exit_code = 1
+        summary_payload = None
+
+        try:
+            summary_exit_code, summary_payload, _ = run_loop_payload(
+                extracted_payload=extracted_payload,
+                prompt_path=cfg.prompt_path,
+                max_retries=cfg.max_retries,
+                timeout_seconds=cfg.timeout_seconds,
+                api_key=resolved_api_key,
+                model=resolved_model,
+                api_url=resolved_api_url,
+                output_path=None,
+                validator=validate_article_summary_payload,
+            )
+        except httpx.TimeoutException:
+            summary_exit_code = 1
+            summary_payload = None
+
+        should_fallback = (
+            (summary_exit_code != 0 or summary_payload is None)
+            and resolved_model is not None
+            and resolved_model == (cfg.llm_model or model or resolved_model)
+            and resolved_model.startswith("deepseek")
         )
+
+        if should_fallback:
+            summary_exit_code, summary_payload, _ = run_loop_payload(
+                extracted_payload=extracted_payload,
+                prompt_path=cfg.prompt_path,
+                max_retries=cfg.max_retries,
+                timeout_seconds=cfg.timeout_seconds,
+                api_key=None,
+                model=None,
+                api_url=None,
+                output_path=None,
+                validator=validate_article_summary_payload,
+            )
+
         if summary_exit_code != 0 or summary_payload is None:
             continue