Spiral gives your agent every tool with one account. Browse the tools →

Codex · ImageReleased 2025-09-28

How to Use HunyuanImage 3.0 in Codex

HunyuanImage 3.0 turns detailed text briefs into photorealistic scenes, illustrations, and graphics with lettering. Its original checkpoint benefits from an explicit visual description because it does not rewrite prompts itself.

Prompts to try with HunyuanImage 3.0 in Codex

Copy an example into Codex, or ask in your own words.

Read the homepage copy and brand notes in this project. Use HunyuanImage 3.0 to create a 1024 × 1024 editorial illustration of the main customer problem, with one clear subject and action, the project's colors, and no lettering.

HunyuanImage 3.0Codex
Try asking
  • Use HunyuanImage 3.0 to make a static design study for this project's sign-in screen. Read the existing labels, describe each region from top to bottom, specify exactly two input boxes and one submit button, and put every visible label in double quotes.

  • Read the product concept in README.md. Use HunyuanImage 3.0 to create a 1024 × 1024 product photograph concept, explicitly describing the object's materials, a three-quarter camera angle, and soft light from the left.

  • Review the draft article in this project and choose one concrete visual metaphor for its argument. Use HunyuanImage 3.0 to create a landscape illustration at 1280 × 768, placing the main subject on the right and leaving the left side clear.

  • Read the event brief in this project. Use HunyuanImage 3.0 to draft a square poster with the exact event title in double quotes, a clearly described central scene, and an explicit position for the title; list the lettering I should check.

How to set up HunyuanImage 3.0 in Codex

Pick the way that suits you. Each takes about two minutes.

Use Spiral's MCP server

  1. Add the hosted server
    codex mcp add spiral --url https://api.spiral.ai/mcp
  2. Sign in

    Codex opens the browser: sign in to Spiral and click Authorize. No API key. codex mcp list then shows Spiral.

    codex mcp login spiral

Use the Spiral skill

  1. Send Codex this message

    Your agent reads Spiral's skill and sets itself up.

    set up https://spiral.ai/SKILL.md
  2. Sign in or paste a key

    Approve Spiral in the browser, or give the agent a key from the console.

Use the Spiral CLI

  1. Install and sign in

    Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

    npx -y spiral-cli@latest init

Call the Spiral API

  1. Create an API key

    In the console, under API keys.

  2. Run HunyuanImage 3.0

    One request; the result comes back by polling or a webhook.

    curl https://api.spiral.ai/v1/runs \
      -H "Authorization: Bearer $SPIRAL_API_KEY" \
      -H "Content-Type: application/json" \
      -H "Idempotency-Key: $(uuidgen)" \
      -d '{
        "provider": "fal",
        "model": "fal-ai/hunyuan-image/v3/text-to-image",
        "input": {
          "prompt": "200mm telephoto through crowd gaps; subject laughing, candid; creamy background compression, color pop from a single bold garment, catchlight in eyes."
        }
      }'

Then paste one of the prompts above: Codex finds HunyuanImage 3.0 on Spiral, shows the price and runs it.

HunyuanImage 3.0 providers and prices

Spiral runs HunyuanImage 3.0 through fal. Open a card for today's price, with every fee included.

falImage
Hunyuan Image
fal-ai/hunyuan-image/v3/text-to-image

Settings you can ask for

HunyuanImage 3.0 works from your prompt. Ask Codex for any of these in your prompt.

SettingAllowed valuesDefaultDescription
Negative promptwhat to leave out, in words

The negative prompt to guide the image generation away from certain concepts.

Number of images1 to 41

The number of images to generate.

Number of inference steps1 to 5028

Number of denoising steps.

Guidance scale1 to 207.5

Controls how much the model adheres to the prompt. Higher values mean stricter adherence.

Seeda whole number

Random seed for reproducible results. If None, a random seed is used.

Output formatjpeg or pngpng

The format of the generated image.

Enable prompt expansionon or offoff

Whether to enable prompt expansion. This will use a large language model to expand the prompt with additional details while maintaining the original meaning.

Every API field of Hunyuan Image is on its tool page: open its card above.

What can you do with HunyuanImage 3.0 in Codex?

Create visual assets and static design references from the project's written requirements. Let the coding agent turn those requirements into explicit subjects, layouts, and lettering before generating an image.

Article illustrations

Turn an article's central idea into a concrete scene. Describe the subject, action, setting, and composition, with open space for a headline added later.

Product concept photography

Explore a proposed product's shape, materials, setting, and lighting from a written brief. Use the images to discuss a concept before arranging a shoot.

Event poster drafts

Develop a poster around a specific scene and a short quoted title. State where each text element belongs and check the lettering before production.

Interface design studies

Explore a static screen's visual direction with explicit regions, element counts, colors, and quoted labels. Use the result as a design reference during implementation.

Campaign art direction

Compare visual treatments of the same written campaign idea by changing the setting, camera angle, or lighting in separate generations.

Presentation cover images

Translate an abstract presentation topic into a specific visual metaphor, such as a repaired bridge for restoring trust. Describe the metaphor's visible objects and their relationships.

Questions about HunyuanImage 3.0 in Codex

Can Codex use HunyuanImage 3.0?

Yes. Add Spiral to Codex as an MCP server, and Codex can find and run HunyuanImage 3.0 like any built-in tool.

How much does HunyuanImage 3.0 cost in Codex?

$0.13 per megapixel. `image_size` presets count as 1 MP; custom sizes round up to whole megapixels. Billed on `num_images`, default 1. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.

Do I need a fal account to use HunyuanImage 3.0 in Codex?

No. Spiral runs HunyuanImage 3.0 on its own fal account. You sign in to Spiral once and never paste a key.

How long does it take to set up HunyuanImage 3.0 in Codex?

About two minutes: add the hosted server, then sign in.

Does Codex wait while HunyuanImage 3.0 runs?

HunyuanImage 3.0 runs in the background. Codex starts the run, checks on it and gives you the image when it's ready.

What is HunyuanImage 3.0 good at?

Detailed photorealistic scenes, concept illustrations, and graphics containing text are useful starting points. Tencent highlights prompt adherence, concept interpretation, and text rendering as strengths.

How should I write a HunyuanImage 3.0 prompt?

Start with the subject and action, then add the setting and style. Specify the framing, positions, and lighting you want instead of asking for a generally attractive image.

Can HunyuanImage 3.0 put words inside an image?

Yes, text rendering is one of Tencent's stated strengths. Put the exact words in double quotes, describe their position, and proofread the result before using it.

Which languages should I use for prompts?

English and Chinese are supported by Tencent's bilingual training descriptions. For lettering, write the exact text you want to appear and keep its original language.

What image sizes does Tencent document?

Tencent demonstrates 1024 × 1024 generation and documents requests such as 1280 × 768 and 16:9. Those are documented examples, rather than a verified maximum resolution.

Get started

Start using HunyuanImage 3.0 in Codex

One Spiral account for every tool your agent needs. No subscription; you pay per run.

Paste this into Codex:

Set up Codex

set up https://spiral.ai/SKILL.md
Are you an AI agent? See setup options