diff --git a/tests/tools/test_hunyuan_image.py b/tests/tools/test_hunyuan_image.py
new file mode 100644
index 00000000..efb1f9d5
--- /dev/null
+++ b/tests/tools/test_hunyuan_image.py
@@ -0,0 +1,488 @@
+"""Unit tests for hunyuan_image — TokenHub 混元生图 3.0 tool."""
+
+from __future__ import annotations
+
+import json
+import sys
+from pathlib import Path
+from unittest.mock import patch
+
+import pytest
+
+PROJECT_ROOT = Path(__file__).resolve().parent.parent.parent
+sys.path.insert(0, str(PROJECT_ROOT))
+
+from tools.base_tool import ToolStatus
+
+# ---------------------------------------------------------------------------
+# Tool discovery & metadata
+# ---------------------------------------------------------------------------
+
+
+def test_hunyuan_image_is_discovered_by_registry():
+    from tools.tool_registry import ToolRegistry
+
+    registry = ToolRegistry()
+    registry.discover()
+
+    tool = registry.get("hunyuan_image")
+    assert tool is not None
+    assert tool.provider == "hunyuan_cloud"
+    assert tool.capability == "image_generation"
+    assert tool.name == "hunyuan_image"
+
+
+def test_hunyuan_image_metadata():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    tool = HunyuanImage()
+    info = tool.get_info()
+
+    assert info["tier"] == "generate"
+    assert info["stability"] == "experimental"
+    assert info["runtime"] == "api"
+    assert "text_to_image" in info["capabilities"]
+    assert info["supports"]["seed"] is True
+    assert info["supports"]["reference_image"] is True
+    assert info["supports"]["prompt_rewrite"] is True
+    assert info["supports"]["negative_prompt"] is False
+
+
+# ---------------------------------------------------------------------------
+# Status reporting
+# ---------------------------------------------------------------------------
+
+
+def test_status_unavailable_when_no_api_key(monkeypatch):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    monkeypatch.delenv("TENCENT_TOKENHUB_API_KEY", raising=False)
+    assert HunyuanImage().get_status() == ToolStatus.UNAVAILABLE
+
+
+def test_status_available_when_api_key_set(monkeypatch):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    monkeypatch.setenv("TENCENT_TOKENHUB_API_KEY", "test-key-123")
+    assert HunyuanImage().get_status() == ToolStatus.AVAILABLE
+
+
+def test_api_key_filters_comment_like_values(monkeypatch):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    monkeypatch.setenv("TENCENT_TOKENHUB_API_KEY", "# this-is-a-comment")
+    assert HunyuanImage()._api_key() is None
+    assert HunyuanImage().get_status() == ToolStatus.UNAVAILABLE
+
+
+def test_api_key_strips_whitespace(monkeypatch):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    monkeypatch.setenv("TENCENT_TOKENHUB_API_KEY", "  my-key  ")
+    assert HunyuanImage()._api_key() == "my-key"
+
+
+# ---------------------------------------------------------------------------
+# Cost & runtime estimation
+# ---------------------------------------------------------------------------
+
+
+def test_estimate_cost():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    tool = HunyuanImage()
+    cost = tool.estimate_cost({})
+    assert cost > 0
+    assert cost == pytest.approx(0.08, rel=0.1)
+
+
+def test_estimate_runtime():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    tool = HunyuanImage()
+    runtime = tool.estimate_runtime({})
+    assert runtime == 120.0
+
+
+# ---------------------------------------------------------------------------
+# Payload construction (_build_payload)
+# ---------------------------------------------------------------------------
+
+
+def test_build_payload_minimal():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    tool = HunyuanImage()
+    payload = tool._build_payload({"prompt": "a cat"})
+    assert payload == {"prompt": "a cat"}
+
+
+def test_build_payload_all_params():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    tool = HunyuanImage()
+    payload = tool._build_payload({
+        "prompt": "a cat",
+        "resolution": "768:1024",
+        "seed": 42,
+        "revise": 0,
+        "logo_add": 1,
+        "images": ["https://example.com/ref.jpg"],
+    })
+    assert payload["prompt"] == "a cat"
+    assert payload["resolution"] == "768:1024"
+    assert payload["seed"] == 42
+    assert payload["revise"] == 0
+    assert payload["logo_add"] == 1
+    assert payload["images"] == ["https://example.com/ref.jpg"]
+
+
+def test_build_payload_with_logo_param():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    tool = HunyuanImage()
+    payload = tool._build_payload({
+        "prompt": "a cat",
+        "logo_param": {"logo_url": "https://example.com/wm.png"},
+    })
+    assert payload["logo_param"] == {"logo_url": "https://example.com/wm.png"}
+
+
+def test_build_payload_logo_param_skips_empty():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    tool = HunyuanImage()
+    payload = tool._build_payload({
+        "prompt": "a cat",
+        "logo_param": {},
+    })
+    assert "logo_param" not in payload
+
+
+def test_build_payload_omits_none_seed():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    tool = HunyuanImage()
+    payload = tool._build_payload({"prompt": "a cat", "seed": None})
+    assert "seed" not in payload
+
+
+# ---------------------------------------------------------------------------
+# Image resolution (_resolve_images)
+# ---------------------------------------------------------------------------
+
+
+def test_resolve_images_passes_urls_through():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    refs = [
+        "https://example.com/a.jpg",
+        "data:image/png;base64,abc123",
+    ]
+    resolved = HunyuanImage._resolve_images(refs)
+    assert resolved == refs
+
+
+def test_resolve_images_encodes_local_file(tmp_path):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    img = tmp_path / "test.png"
+    img.write_bytes(b"fake-png-data")
+
+    resolved = HunyuanImage._resolve_images([str(img)])
+    assert len(resolved) == 1
+    assert resolved[0].startswith("data:image/png;base64,")
+
+
+def test_resolve_images_raises_on_missing_file():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    with pytest.raises(FileNotFoundError):
+        HunyuanImage._resolve_images(["/nonexistent/path.jpg"])
+
+
+def test_resolve_images_raises_on_oversized_file(tmp_path):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    big = tmp_path / "big.jpg"
+    big.write_bytes(b"x" * (7 * 1024 * 1024))  # 7MB > 6MB limit
+
+    with pytest.raises(ValueError, match="too large"):
+        HunyuanImage._resolve_images([str(big)])
+
+
+def test_resolve_images_detects_mime_from_extension(tmp_path):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    cases = [
+        ("ref.jpg", "image/jpeg"),
+        ("ref.jpeg", "image/jpeg"),
+        ("ref.png", "image/png"),
+        ("ref.bmp", "image/bmp"),
+        ("ref.tiff", "image/tiff"),
+        ("ref.tif", "image/tiff"),
+        ("ref.webp", "image/webp"),
+        ("ref.unknown", "image/png"),  # fallback
+    ]
+    for filename, expected_mime in cases:
+        f = tmp_path / filename
+        f.write_bytes(b"data")
+        resolved = HunyuanImage._resolve_images([str(f)])
+        assert resolved[0].startswith(f"data:{expected_mime};base64,")
+
+
+# ---------------------------------------------------------------------------
+# Output path resolution (_resolve_output_paths)
+# ---------------------------------------------------------------------------
+
+
+def test_resolve_output_paths_single():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    paths = HunyuanImage._resolve_output_paths("/out/img.png", 1)
+    assert len(paths) == 1
+    assert paths[0] == Path("/out/img.png")
+
+
+def test_resolve_output_paths_multi():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    paths = HunyuanImage._resolve_output_paths("/out/img.png", 3)
+    assert len(paths) == 3
+    assert [p.name for p in paths] == ["img_1.png", "img_2.png", "img_3.png"]
+    assert len(set(paths)) == 3
+
+
+# ---------------------------------------------------------------------------
+# Auth headers
+# ---------------------------------------------------------------------------
+
+
+def test_auth_headers():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    headers = HunyuanImage._auth_headers("my-api-key")
+    assert headers["Authorization"] == "Bearer my-api-key"
+    assert headers["Content-Type"] == "application/json"
+
+
+# ---------------------------------------------------------------------------
+# JSON error handling
+# ---------------------------------------------------------------------------
+
+
+def test_json_or_raise_parses_valid_json():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    class FakeResp:
+        status_code = 200
+
+        def json(self):
+            return {"status": "ok"}
+
+    assert HunyuanImage._json_or_raise(FakeResp()) == {"status": "ok"}
+
+
+def test_json_or_raise_raises_on_invalid_json():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    class FakeResp:
+        status_code = 500
+
+        def json(self):
+            raise ValueError("not json")
+
+    with pytest.raises(RuntimeError, match="Non-JSON response"):
+        HunyuanImage._json_or_raise(FakeResp())
+
+
+def test_check_response_passes_clean_payload():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    HunyuanImage._check_response({"status": "completed"})  # no error -> no raise
+
+
+def test_check_response_raises_on_error_field():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    with pytest.raises(RuntimeError, match="TokenHub API error"):
+        HunyuanImage._check_response({
+            "error": {"code": "AUTH_FAILED", "message": "invalid key"},
+        })
+
+
+def test_check_response_raises_on_error_without_code():
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    with pytest.raises(RuntimeError, match="TokenHub API error"):
+        HunyuanImage._check_response({"error": {"message": "something broke"}})
+
+
+# ---------------------------------------------------------------------------
+# Safe error redaction
+# ---------------------------------------------------------------------------
+
+
+def test_safe_error_redacts_api_key(monkeypatch):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    monkeypatch.setenv("TENCENT_TOKENHUB_API_KEY", "secret-key-abc")
+    msg = HunyuanImage._safe_error(Exception("failed with secret-key-abc"))
+    assert "secret-key-abc" not in msg
+    assert "[redacted]" in msg
+
+
+def test_safe_error_preserves_other_text(monkeypatch):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    monkeypatch.setenv("TENCENT_TOKENHUB_API_KEY", "sk-123")
+    msg = HunyuanImage._safe_error(Exception("network timeout: connection refused"))
+    assert "network timeout" in msg
+    assert "sk-123" not in msg
+
+
+# ---------------------------------------------------------------------------
+# Execute guards
+# ---------------------------------------------------------------------------
+
+
+def test_execute_returns_error_without_api_key(monkeypatch):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    monkeypatch.delenv("TENCENT_TOKENHUB_API_KEY", raising=False)
+    result = HunyuanImage().execute({"prompt": "a cat"})
+    assert not result.success
+    assert "TENCENT_TOKENHUB_API_KEY" in result.error
+
+
+# ---------------------------------------------------------------------------
+# Dry run
+# ---------------------------------------------------------------------------
+
+
+def test_dry_run_no_side_effects(monkeypatch):
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    monkeypatch.setenv("TENCENT_TOKENHUB_API_KEY", "test-key")
+    tool = HunyuanImage()
+    info = tool.dry_run({"prompt": "a cat"})
+    assert info["tool"] == "hunyuan_image"
+    assert info["estimated_cost_usd"] > 0
+    assert info["would_execute"] is True
+
+
+# ---------------------------------------------------------------------------
+# End-to-end with mocked API
+# ---------------------------------------------------------------------------
+
+
+def test_execute_full_flow_with_mocked_api(monkeypatch, tmp_path):
+    """Simulate the full submit → poll → download flow."""
+    from tools.graphics.hunyuan_image import HunyuanImage, _MODEL, _HOST
+
+    monkeypatch.setenv("TENCENT_TOKENHUB_API_KEY", "test-key")
+
+    # Track calls across all mocked endpoints
+    api_calls = []
+    poll_count = [0]  # mutable counter for poll iteration
+
+    class _FakeResp:
+        status_code = 200
+        def __init__(self, data, content=None):
+            self._data = data
+            self.content = content
+        def json(self):
+            return self._data
+        def raise_for_status(self):
+            pass
+
+    submit_url = f"https://{_HOST}/v1/api/image/submit"
+    query_url = f"https://{_HOST}/v1/api/image/query"
+
+    # Safety sentinel: if a real network call slips through the mock,
+    # this flag is flipped and we fail-fast instead of leaking files.
+    mock_active = [False]
+
+    def fake_post(url, *, json, headers, timeout):
+        mock_active[0] = True
+        api_calls.append(("post", url, json))
+        if "submit" in url:
+            return _FakeResp({"id": "job-001", "status": "queued"})
+        elif "query" in url:
+            poll_count[0] += 1
+            if poll_count[0] == 1:
+                return _FakeResp({"status": "running"})
+            return _FakeResp({
+                "status": "completed",
+                "data": [{"url": "https://example.com/result.png"}],
+            })
+        raise RuntimeError(f"Unexpected URL: {url}")
+
+    def fake_get(url, timeout):
+        mock_active[0] = True
+        api_calls.append(("get", url))
+        return _FakeResp({}, content=b"fake-image-data")
+
+    with (
+        patch("requests.post", side_effect=fake_post),
+        patch("requests.get", side_effect=fake_get),
+    ):
+        out = tmp_path / "gen.png"
+        result = HunyuanImage().execute({
+            "prompt": "a programmer coding",
+            "resolution": "1024:1024",
+            "seed": 12345,
+            "revise": 1,
+            "logo_add": 0,
+            "output_path": str(out),
+        })
+
+    # Guard: mock must have been exercised — if not, a real API call leaked
+    assert mock_active[0], (
+        "Mock was never triggered — a real API call may have leaked. "
+        "Check that requests.post / requests.get patching is effective."
+    )
+
+    assert result.success, result.error
+    assert result.data["provider"] == "hunyuan_cloud"
+    assert result.data["model"] == _MODEL
+    assert result.data["task_id"] == "job-001"
+    assert result.data["resolution"] == "1024:1024"
+    assert result.data["images_generated"] == 1
+    assert result.artifacts == [str(out)]
+    assert out.read_bytes() == b"fake-image-data"
+
+    # Verify submit payload was correct
+    submit_calls = [c for c in api_calls if "submit" in c[1]]
+    assert len(submit_calls) == 1
+    _, _, submit_body = submit_calls[0]
+    assert submit_body["prompt"] == "a programmer coding"
+    assert submit_body["resolution"] == "1024:1024"
+    assert submit_body["seed"] == 12345
+    assert submit_body["revise"] == 1
+    assert submit_body["logo_add"] == 0
+    assert submit_body["model"] == _MODEL
+
+    # Verify polling happened (submit + at least 1 query + download)
+    assert any("query" in c[1] for c in api_calls)
+    assert any(c[0] == "get" for c in api_calls)
+
+
+def test_execute_with_local_reference_images(monkeypatch, tmp_path):
+    """Reference images from local paths should be base64-encoded in payload."""
+    from tools.graphics.hunyuan_image import HunyuanImage
+
+    monkeypatch.setenv("TENCENT_TOKENHUB_API_KEY", "test-key")
+
+    ref_img = tmp_path / "ref.png"
+    ref_img.write_bytes(b"reference-data")
+
+    tool = HunyuanImage()
+    payload = tool._build_payload({
+        "prompt": "enhance this",
+        "images": [str(ref_img)],
+    })
+
+    assert "images" in payload
+    assert len(payload["images"]) == 1
+    assert payload["images"][0].startswith("data:image/png;base64,")
diff --git a/tools/graphics/hunyuan_image.py b/tools/graphics/hunyuan_image.py
new file mode 100644
index 00000000..65a52c8c
--- /dev/null
+++ b/tools/graphics/hunyuan_image.py
@@ -0,0 +1,579 @@
+"""Tencent Hunyuan (腾讯混元) cloud image generation (3.0) via TokenHub API.
+
+Calls the Tencent TokenHub API (tokenhub.tencentmaas.com) using simple Bearer
+token authentication. This is the OpenAI-compatible API gateway for Tencent
+Hunyuan image models — no TC3-HMAC-SHA256 signing required.
+
+API flow: POST /v1/api/image/submit -> poll /v1/api/image/query ->
+download data[].url.
+
+Authentication uses a TokenHub API key obtained from the Tencent Cloud
+TokenHub console (https://console.cloud.tencent.com/tokenhub).
+"""
+
+from __future__ import annotations
+
+import os
+import time
+from pathlib import Path
+from typing import Any
+
+from tools.base_tool import (
+    BaseTool,
+    Determinism,
+    ExecutionMode,
+    ResourceProfile,
+    RetryPolicy,
+    ToolResult,
+    ToolRuntime,
+    ToolStability,
+    ToolStatus,
+    ToolTier,
+)
+
+_HOST = "tokenhub.tencentmaas.com"
+_SUBMIT_PATH = "/v1/api/image/submit"
+_QUERY_PATH = "/v1/api/image/query"
+
+# TokenHub model identifier for 混元生图 3.0
+_MODEL = "hy-image-v3.0"
+
+
+class HunyuanImage(BaseTool):
+    """Tencent Hunyuan cloud image generation (3.0) via TokenHub API."""
+
+    name = "hunyuan_image"
+    version = "0.1.0"
+    tier = ToolTier.GENERATE
+    capability = "image_generation"
+    provider = "hunyuan_cloud"
+    stability = ToolStability.EXPERIMENTAL
+    execution_mode = ExecutionMode.ASYNC
+    determinism = Determinism.SEEDED
+    runtime = ToolRuntime.API
+
+    dependencies = []
+    install_instructions = (
+        "Set TENCENT_TOKENHUB_API_KEY to your Tencent Cloud TokenHub API key.\n"
+        "  Get it at https://console.cloud.tencent.com/tokenhub"
+    )
+    agent_skills = []
+
+    capabilities = ["generate_image", "text_to_image"]
+    supports = {
+        "negative_prompt": False,
+        "seed": True,
+        "custom_size": True,
+        "reference_image": True,
+        "prompt_rewrite": True,
+    }
+    best_for = [
+        "Hunyuan text-to-image via Tencent TokenHub API",
+        "simple Bearer-token auth (no TC3 signing required)",
+        "direct Tencent Cloud quota usage (not through a third-party gateway)",
+        "Chinese-language prompt understanding",
+    ]
+    not_good_for = [
+        "offline generation or air-gapped environments",
+        "users without Tencent Cloud account and real-name verification",
+    ]
+    fallback_tools = ["dashscope_image", "flux_image", "openai_image", "recraft_image"]
+
+    input_schema = {
+        "type": "object",
+        "required": ["prompt"],
+        "properties": {
+            "prompt": {
+                "type": "string",
+                "maxLength": 8192,
+                "description": (
+                    "Image description. Max 8192 UTF-8 characters. "
+                    "Supports Chinese and English. Be specific about subject, "
+                    "composition, style, and mood."
+                ),
+            },
+            "images": {
+                "type": "array",
+                "items": {"type": "string"},
+                "maxItems": 3,
+                "description": (
+                    "Reference images per upstream Images.N param (max 3). "
+                    "Each entry is a publicly accessible URL or a local file path "
+                    "(auto-encoded to base64 data URI). "
+                    "Single image: 50-5000px per side, base64 < 6MB. "
+                    "Formats: jpg/png/jpeg/webp/bmp/tiff."
+                ),
+            },
+            "resolution": {
+                "type": "string",
+                "default": "1024:1024",
+                "description": (
+                    'Image resolution as "W:H" (colon separator, per upstream '
+                    "Resolution param). W, H in [512, 2048], product (W*H) <= "
+                    '1024x1024 pixels. Examples: "1024:1024", "768:1024", '
+                    '"1024:576".'
+                ),
+            },
+            "seed": {
+                "type": "integer",
+                "minimum": 1,
+                "maximum": 4294967295,
+                "description": (
+                    "Random seed in [1, 4294967295]. "
+                    "Note: seed is ignored when revise is enabled (default)."
+                ),
+            },
+            "revise": {
+                "type": "integer",
+                "enum": [0, 1],
+                "default": 1,
+                "description": (
+                    "Prompt auto-rewrite toggle per upstream Revise param. "
+                    "1 = enabled (default, adds ~20s processing), 0 = disabled. "
+                    "When disabled, caller should handle prompt rewriting."
+                ),
+            },
+            "logo_add": {
+                "type": "integer",
+                "enum": [0, 1],
+                "default": 1,
+                "description": (
+                    "Add 'AI-generated' watermark per upstream LogoAdd param. "
+                    "1 = add watermark (default), 0 = no watermark. "
+                    "Values other than 0 or 1 are treated as 1."
+                ),
+            },
+            "logo_param": {
+                "type": "object",
+                "properties": {
+                    "logo_url": {
+                        "type": "string",
+                        "description": "Custom watermark image URL.",
+                    },
+                    "logo_image": {
+                        "type": "string",
+                        "description": "Custom watermark image as base64-encoded string.",
+                    },
+                },
+                "description": (
+                    "Custom watermark settings per upstream LogoParam. "
+                    "Default: \"图片由 AI 生成\" at bottom-right. "
+                    "Note: custom watermarks may not be supported by the engine "
+                    "(error: InvalidParameterValue.LogoParamErr)."
+                ),
+            },
+            "output_path": {
+                "type": "string",
+                "description": "Output file path for the generated image (PNG).",
+            },
+            "poll_interval_seconds": {
+                "type": "number",
+                "minimum": 2,
+                "default": 5.0,
+                "description": "Seconds between status polls.",
+            },
+            "timeout_seconds": {
+                "type": "integer",
+                "minimum": 60,
+                "default": 600,
+                "description": "Maximum seconds to wait for generation.",
+            },
+        },
+    }
+
+    resource_profile = ResourceProfile(
+        cpu_cores=1, ram_mb=512, vram_mb=0, disk_mb=100, network_required=True,
+    )
+    retry_policy = RetryPolicy(
+        max_retries=2,
+        backoff_seconds=2.0,
+        retryable_errors=["rate_limit", "timeout"],
+    )
+    idempotency_key_fields = [
+        "prompt",
+        "resolution",
+        "images",
+        "seed",
+        "revise",
+        "logo_add",
+    ]
+    side_effects = [
+        "writes image file to output_path",
+        "calls Tencent TokenHub API (Bearer-token submit + poll + download)",
+    ]
+    user_visible_verification = [
+        "Inspect generated image for quality and prompt adherence",
+        "Check for watermark if logo_add=0 was requested",
+    ]
+
+    # ------------------------------------------------------------------
+    # Credential helpers
+    # ------------------------------------------------------------------
+
+    @staticmethod
+    def _api_key() -> str | None:
+        val = os.environ.get("TENCENT_TOKENHUB_API_KEY", "")
+        if val and not val.strip().startswith("#"):
+            return val.strip()
+        return None
+
+    # ------------------------------------------------------------------
+    # Tool contract methods
+    # ------------------------------------------------------------------
+
+    def get_status(self) -> ToolStatus:
+        if self._api_key():
+            return ToolStatus.AVAILABLE
+        return ToolStatus.UNAVAILABLE
+
+    def estimate_cost(self, inputs: dict[str, Any]) -> float:
+        """Estimate cost in USD.
+
+        Tencent TokenHub credit-based pricing (1 credit = 1.2 RMB ≈ $0.167 USD):
+        - hy-image-v3.0: ~0.5 credits/image → ~$0.08
+
+        Source: https://cloud.tencent.com.cn/document/product/1823/130054
+        """
+        _CREDIT_TO_USD = 1.2 / 7.2  # 1 credit = 1.2 RMB, ~7.2 RMB/USD
+        credits = 0.5
+        return round(credits * _CREDIT_TO_USD, 2)
+
+    def estimate_runtime(self, inputs: dict[str, Any]) -> float:
+        """Estimate wall-clock time in seconds.
+
+        Per upstream docs, prompt rewrite (revise=1) adds ~20s. Including
+        queuing and download, 120s is a safe upper-bound.
+        """
+        return 120.0
+
+    # ------------------------------------------------------------------
+    # Main execution
+    # ------------------------------------------------------------------
+
+    def execute(self, inputs: dict[str, Any]) -> ToolResult:
+        api_key = self._api_key()
+        if not api_key:
+            return ToolResult(
+                success=False,
+                error="TENCENT_TOKENHUB_API_KEY not set. " + self.install_instructions,
+            )
+
+        start = time.time()
+        try:
+            result = self._generate(inputs, api_key=api_key)
+        except Exception as exc:
+            return ToolResult(
+                success=False,
+                error=f"Hunyuan TokenHub image generation failed: {self._safe_error(exc)}",
+            )
+
+        result.duration_seconds = round(time.time() - start, 2)
+        return result
+
+    # ------------------------------------------------------------------
+    # Generation pipeline
+    # ------------------------------------------------------------------
+
+    def _generate(
+        self, inputs: dict[str, Any], *, api_key: str,
+    ) -> ToolResult:
+        import requests
+
+        # Guard: refuse to make paid API calls without an explicit output_path.
+        # A CWD-relative default would leak files into the project root when
+        # called by selectors or other automated tooling.
+        if not inputs.get("output_path"):
+            return ToolResult(
+                success=False,
+                error="output_path is required for hunyuan_image generation.",
+            )
+
+        payload = self._build_payload(inputs)
+        task_id = self._submit_task(payload, model=_MODEL, api_key=api_key)
+        image_urls = self._poll_task(
+            task_id,
+            model=_MODEL,
+            api_key=api_key,
+            poll_interval=float(inputs.get("poll_interval_seconds", 5.0)),
+            timeout_seconds=int(inputs.get("timeout_seconds", 600)),
+        )
+
+        output_paths = self._resolve_output_paths(
+            inputs["output_path"],
+            count=len(image_urls),
+        )
+        for path, url in zip(output_paths, image_urls):
+            download = requests.get(url, timeout=120)
+            download.raise_for_status()
+            path.parent.mkdir(parents=True, exist_ok=True)
+            path.write_bytes(download.content)
+
+        return ToolResult(
+            success=True,
+            data={
+                "provider": "hunyuan_cloud",
+                "route": "tokenhub",
+                "model": _MODEL,
+                "prompt": inputs["prompt"],
+                "resolution": payload.get("resolution", "1024:1024"),
+                "revise": payload.get("revise", 1),
+                "logo_add": payload.get("logo_add", 1),
+                "task_id": task_id,
+                "output": str(output_paths[0]),
+                "outputs": [str(p) for p in output_paths],
+                "images_generated": len(output_paths),
+            },
+            artifacts=[str(p) for p in output_paths],
+            cost_usd=self.estimate_cost(inputs),
+            model=_MODEL,
+        )
+
+    # ------------------------------------------------------------------
+    # Payload construction
+    # ------------------------------------------------------------------
+
+    def _build_payload(self, inputs: dict[str, Any]) -> dict[str, Any]:
+        """Build the request body for TokenHub image submit.
+
+        TokenHub async endpoints use snake_case versions of the upstream
+        SubmitTextToImageJob parameter names (e.g. Resolution -> resolution,
+        Images -> images).
+        """
+        payload: dict[str, Any] = {
+            "prompt": inputs["prompt"],
+        }
+
+        # Optional parameters (snake_case of upstream SubmitTextToImageJob params)
+        if inputs.get("resolution"):
+            payload["resolution"] = inputs["resolution"]
+        if inputs.get("seed") is not None:
+            payload["seed"] = int(inputs["seed"])
+        if "revise" in inputs:
+            payload["revise"] = int(inputs["revise"])
+        if "logo_add" in inputs:
+            payload["logo_add"] = int(inputs["logo_add"])
+        if inputs.get("logo_param"):
+            logo_param: dict[str, str] = {}
+            lp = inputs["logo_param"]
+            if lp.get("logo_url"):
+                logo_param["logo_url"] = lp["logo_url"]
+            if lp.get("logo_image"):
+                logo_param["logo_image"] = lp["logo_image"]
+            if logo_param:
+                payload["logo_param"] = logo_param
+
+        # Reference images — maps to upstream Images.N
+        # TokenHub accepts URLs or base64 data URIs in the images array
+        image_refs = inputs.get("images")
+        if image_refs:
+            payload["images"] = self._resolve_images(image_refs)
+
+        return payload
+
+    @staticmethod
+    def _resolve_images(refs: list[str]) -> list[str]:
+        """Resolve reference images to strings for the TokenHub API.
+
+        Each entry may be:
+        - An HTTP(S) URL → passed through unchanged
+        - A data URI (``data:...``) → passed through unchanged
+        - A local file path → base64-encoded as a data URI
+
+        Per upstream docs: single image 50-5000px per side, base64 < 6MB.
+        Formats: jpg/jpeg/png/bmp/tiff/webp.
+        """
+        import base64
+
+        resolved: list[str] = []
+        for ref in refs:
+            if ref.startswith("data:") or ref.startswith("http://") or ref.startswith("https://"):
+                resolved.append(ref)
+                continue
+
+            image_path = Path(ref)
+            if not image_path.is_file():
+                raise FileNotFoundError(f"Reference image not found: {ref}")
+
+            raw = image_path.read_bytes()
+            max_raw = 6 * 1024 * 1024  # 6MB per upstream limit
+            if len(raw) > max_raw:
+                raise ValueError(
+                    f"Image too large ({len(raw)} bytes). Max ~6MB raw."
+                )
+
+            suffix = image_path.suffix.lower()
+            mime_map = {
+                ".jpg": "image/jpeg",
+                ".jpeg": "image/jpeg",
+                ".png": "image/png",
+                ".bmp": "image/bmp",
+                ".tiff": "image/tiff",
+                ".tif": "image/tiff",
+                ".webp": "image/webp",
+            }
+            mime = mime_map.get(suffix, "image/png")
+            data = base64.b64encode(raw).decode("ascii")
+            resolved.append(f"data:{mime};base64,{data}")
+        return resolved
+
+    # ------------------------------------------------------------------
+    # API communication (TokenHub OpenAI-compatible)
+    # ------------------------------------------------------------------
+
+    @staticmethod
+    def _auth_headers(api_key: str) -> dict[str, str]:
+        """Build common request headers for TokenHub API calls."""
+        return {
+            "Authorization": f"Bearer {api_key}",
+            "Content-Type": "application/json",
+        }
+
+    def _submit_task(
+        self, payload: dict[str, Any], *, model: str, api_key: str,
+    ) -> str:
+        """Submit an image generation task and return the task ID.
+
+        POST /v1/api/image/submit
+        Request: {"model": "hy-image-v3.0", "prompt": "...", ...}
+        Response: {"id": "...", "status": "queued", ...}
+        """
+        import requests
+
+        body = {
+            "model": model,
+            **payload,
+        }
+        url = f"https://{_HOST}{_SUBMIT_PATH}"
+        resp = requests.post(
+            url,
+            json=body,
+            headers=self._auth_headers(api_key),
+            timeout=30,
+        )
+        data = self._json_or_raise(resp)
+        self._check_response(data)
+
+        task_id = data.get("id")
+        if not task_id:
+            raise RuntimeError(
+                f"TokenHub submit returned no task id: {data}"
+            )
+        return task_id
+
+    def _poll_task(
+        self,
+        task_id: str,
+        *,
+        model: str,
+        api_key: str,
+        poll_interval: float,
+        timeout_seconds: int,
+    ) -> list[str]:
+        """Poll /v1/api/image/query until completion, return image download URLs.
+
+        Response when completed:
+        {"status": "completed", "data": [{"url": "...", "revised_prompt": "..."}]}
+
+        The data array may contain multiple images. Each URL is valid for ~1 hour.
+        """
+        import requests
+
+        url = f"https://{_HOST}{_QUERY_PATH}"
+
+        deadline = time.time() + timeout_seconds
+        while time.time() < deadline:
+            time.sleep(poll_interval)
+
+            resp = requests.post(
+                url,
+                json={"model": model, "id": task_id},
+                headers=self._auth_headers(api_key),
+                timeout=30,
+            )
+            data = self._json_or_raise(resp)
+            self._check_response(data)
+
+            status = data.get("status", "")
+
+            if status == "completed":
+                result_data = data.get("data") or []
+                urls = [item.get("url") for item in result_data if item.get("url")]
+                if not urls:
+                    raise RuntimeError(
+                        f"TokenHub task {task_id} completed but no data[].url: {data}"
+                    )
+                return urls
+
+            if status == "failed":
+                error_info = data.get("error") or {}
+                error_msg = error_info.get("message", "unknown error")
+                raise RuntimeError(
+                    f"TokenHub task {task_id} failed: {error_msg}"
+                )
+
+            # queued / running / in_progress — continue polling
+            if status not in ("queued", "running", "in_progress"):
+                raise RuntimeError(
+                    f"TokenHub task {task_id} returned unknown status: {status}"
+                )
+
+        raise TimeoutError(
+            f"TokenHub task {task_id} did not finish within {timeout_seconds}s"
+        )
+
+    # ------------------------------------------------------------------
+    # Error handling helpers
+    # ------------------------------------------------------------------
+
+    @staticmethod
+    def _safe_error(exc: Exception) -> str:
+        """Redact secret values from exception messages."""
+        msg = str(exc)
+        for var in ("TENCENT_TOKENHUB_API_KEY",):
+            val = os.environ.get(var, "")
+            if val:
+                msg = msg.replace(val, "[redacted]")
+        return msg
+
+    @staticmethod
+    def _json_or_raise(response: Any) -> dict[str, Any]:
+        """Parse JSON response body or raise with HTTP status."""
+        try:
+            return response.json()
+        except ValueError as exc:
+            raise RuntimeError(
+                f"Non-JSON response from TokenHub API: HTTP {response.status_code}"
+            ) from exc
+
+    @staticmethod
+    def _check_response(payload: dict[str, Any]) -> None:
+        """Check the TokenHub API response for errors.
+
+        TokenHub returns errors at the top level with an ``error`` field.
+        """
+        error = payload.get("error")
+        if error:
+            message = error.get("message", "unknown error")
+            code = error.get("code", error.get("type", "unknown"))
+            raise RuntimeError(
+                f"TokenHub API error: code={code}, message={message}"
+            )
+
+    # ------------------------------------------------------------------
+    # Output helpers
+    # ------------------------------------------------------------------
+
+    @staticmethod
+    def _resolve_output_paths(base: str, count: int) -> list[Path]:
+        """Derive distinct paths for ``count`` images.
+
+        Single image keeps the base path unchanged; multiple images insert an
+        index before the extension (foo.png -> foo_1.png, foo_2.png, ...).
+        """
+        base_path = Path(base)
+        if count <= 1:
+            return [base_path]
+        stem = base_path.stem
+        suffix = base_path.suffix
+        parent = base_path.parent
+        return [parent / f"{stem}_{i}{suffix}" for i in range(1, count + 1)]