diff --git a/README.md b/README.md
index c49fc59..dba275f 100644
--- a/README.md
+++ b/README.md
@@ -320,6 +320,7 @@ These are only needed for **headless / CI extraction** (`graphify extract`). Whe
 | `ANTHROPIC_API_KEY` | Claude (Anthropic) backend | `--backend claude` |
 | `GEMINI_API_KEY` or `GOOGLE_API_KEY` | Google Gemini backend | `--backend gemini` |
 | `OPENAI_API_KEY` | OpenAI or OpenAI-compatible APIs | `--backend openai` |
+| `DEEPSEEK_API_KEY` | DeepSeek backend | `--backend deepseek` |
 | `MOONSHOT_API_KEY` | Kimi Code backend | `--backend kimi` |
 | `OLLAMA_BASE_URL` | Ollama local inference URL | `--backend ollama` (default: `http://localhost:11434`) |
 | `OLLAMA_MODEL` | Ollama model name | `--backend ollama` (default: auto-detect) |
@@ -450,7 +451,7 @@ graphify kiro install / uninstall
 graphify antigravity install / uninstall
 
 graphify extract ./docs                        # headless LLM extraction for CI (no IDE needed)
-graphify extract ./docs --backend gemini       # explicit backend: gemini, kimi, claude, openai, ollama, bedrock, or claude-cli
+graphify extract ./docs --backend gemini       # explicit backend: gemini, kimi, claude, openai, deepseek, ollama, bedrock, or claude-cli
 graphify extract ./docs --backend gemini --model gemini-3.1-pro-preview
 graphify extract ./docs --backend ollama       # local Ollama (set OLLAMA_BASE_URL / OLLAMA_MODEL) - no API key needed for loopback
 GRAPHIFY_OLLAMA_NUM_CTX=32768 graphify extract ./docs --backend ollama   # override KV-cache window (auto-sized by default)
diff --git a/graphify/__main__.py b/graphify/__main__.py
index 4fff852..51c4db9 100644
--- a/graphify/__main__.py
+++ b/graphify/__main__.py
@@ -1255,7 +1255,7 @@ def main() -> None:
         print("    --top-k-edges N         per-symbol outbound edges in inspector (default 12)")
         print("    --label NAME            project label in header")
         print("  extract <path>          headless full extraction (AST + semantic LLM) for CI/scripts")
-        print("    --backend B             gemini|kimi|claude|openai|ollama (default: whichever API key is set)")
+        print("    --backend B             gemini|kimi|claude|openai|deepseek|ollama (default: whichever API key is set)")
         print("    --model M               override backend default model")
         print("    --max-workers N         AST extraction subprocess count (default: cpu_count)")
         print("    --token-budget N        per-chunk token cap for semantic extraction (default: 60000)")
@@ -1880,6 +1880,7 @@ def main() -> None:
                 os.environ.get("GEMINI_API_KEY")
                 or os.environ.get("GOOGLE_API_KEY")
                 or os.environ.get("MOONSHOT_API_KEY")
+                or os.environ.get("DEEPSEEK_API_KEY")
                 or os.environ.get("GRAPHIFY_NO_TIPS")
             ):
                 print("Tip: set GEMINI_API_KEY or GOOGLE_API_KEY to use Gemini for semantic extraction.")
@@ -2388,7 +2389,7 @@ def main() -> None:
         # has an API key set.
         if len(sys.argv) < 3:
             print(
-                "Usage: graphify extract <path> [--backend gemini|kimi|claude|openai|ollama] "
+                "Usage: graphify extract <path> [--backend gemini|kimi|claude|openai|deepseek|ollama] "
                 "[--model M] [--out DIR] [--google-workspace] [--no-cluster] "
                 "[--max-workers N] [--token-budget N] [--max-concurrency N] "
                 "[--api-timeout S]",
@@ -2507,7 +2508,8 @@ def main() -> None:
                 print(
                     "error: no LLM API key found. Set GEMINI_API_KEY or GOOGLE_API_KEY "
                     "(gemini), MOONSHOT_API_KEY (kimi), ANTHROPIC_API_KEY (claude), "
-                    "or OPENAI_API_KEY (openai), or pass --backend.",
+                    "OPENAI_API_KEY (openai), DEEPSEEK_API_KEY (deepseek), "
+                    "or pass --backend.",
                     file=sys.stderr,
                 )
                 sys.exit(1)
diff --git a/graphify/llm.py b/graphify/llm.py
index b4392bc..fb65ee1 100644
--- a/graphify/llm.py
+++ b/graphify/llm.py
@@ -87,6 +87,17 @@ BACKENDS: dict[str, dict] = {
         "pricing": {"input": 0.40, "output": 1.60},  # USD per 1M tokens
         "temperature": 0,
     },
+    "deepseek": {
+        "base_url": "https://api.deepseek.com",
+        "default_model": "deepseek-v4-flash",
+        "env_key": "DEEPSEEK_API_KEY",
+        "model_env_key": "GRAPHIFY_DEEPSEEK_MODEL",
+        "pricing": {"input": 0.14, "output": 0.28},  # USD per 1M tokens (v4-flash)
+        # deepseek-reasoner / thinking-mode models silently ignore temperature;
+        # deepseek-chat / v4-flash (non-thinking) accept 0-2. Safe to send 0.
+        "temperature": 0,
+        "max_tokens": 16384,
+    },
     "bedrock": {
         "default_model": "anthropic.claude-3-5-sonnet-20241022-v2:0",
         "model_env_key": "GRAPHIFY_BEDROCK_MODEL",
@@ -1084,7 +1095,7 @@ def detect_backend() -> str | None:
     key now keeps you on the paid backend; remove the paid key (or pass
     --backend ollama explicitly) to route to the local model.
     """
-    for backend in ("gemini", "kimi", "claude", "openai"):
+    for backend in ("gemini", "kimi", "claude", "openai", "deepseek"):
         if _get_backend_api_key(backend):
             return backend
     if os.environ.get("AWS_PROFILE") or os.environ.get("AWS_REGION") or os.environ.get("AWS_DEFAULT_REGION"):