gpt-image-2

Generate and edit images with OpenAI GPT Image 2 (ChatGPT Images 2.0) on RunComfy. Documents GPT Image 2's strengths (embedded text, logos, multilingual typography, instruction precision), its 3 fixed sizes, edit-with-preservation language, and when to route to a sibling (Flux 2 / Nano Banana Pro / Seedream) instead. Calls `runcomfy run openai/gpt-image-2/text-to-image` or `/edit` through the local RunComfy CLI. Triggers on "gpt image 2", "gpt-image-2", "ChatGPT Images 2", "image 2", or any...

npx skills add https://github.com/runcomfy-com/skills --skill gpt-image-2

GPT Image 2 — Pro Pack on RunComfy

runcomfy.com · Text-to-image · Edit · GitHub

OpenAI GPT Image 2 (ChatGPT Images 2.0) hosted on the RunComfy Model API — no OpenAI key, async REST.

npx skills add agentspace-so/runcomfy-skills --skill gpt-image-2 -g

When to pick this model (vs siblings)

GPT Image 2's distinct strength is directive precision: it follows multi-element prompts, layout cues, and embedded-text instructions more reliably than its peers. Pick it when what's on the canvas matters more than how stylized it looks.

You wantUse
Embedded text, logos, signage, multilingual typographyGPT Image 2
Brand-safe, e-commerce / ad / UI mockup imageryGPT Image 2
Iterative refinement that holds composition stableGPT Image 2
Heavy stylization, painterly lookFlux 2
Hyperrealistic portraitNano Banana Pro
Cinematic / aesthetic-first hero shotsSeedream 5

If the user explicitly asked for GPT Image 2 / ChatGPT Image 2 / Image 2, route here regardless — don't second-guess the model choice.

Prerequisites

  1. RunComfy CLInpm i -g @runcomfy/cli
  2. RunComfy accountruncomfy login opens a browser device-code flow.
  3. CI / containers — set RUNCOMFY_TOKEN=<token> instead of runcomfy login.

Endpoints + input schema

Two endpoints, same model.

openai/gpt-image-2/text-to-image

FieldTypeRequiredDefaultNotes
promptstringyesThe positive prompt
sizeenumno1024_10241024_1024 (1:1), 1024_1536 (2:3 portrait), 1536_1024 (3:2 landscape) — only these three

openai/gpt-image-2/edit

FieldTypeRequiredDefaultNotes
promptstringyesNatural-language edit instruction
imagesstring[]yesUp to 10 reference image URLs (publicly fetchable HTTPS)
sizeenumnoautoauto (preserve input ratio), or one of the three fixed sizes above

size=auto on edit preserves the input aspect ratio — strongly recommended unless the edit explicitly changes framing.

How to invoke

Text-to-image:

runcomfy run openai/gpt-image-2/text-to-image \
  --input '{"prompt": "<user prompt>", "size": "1024_1536"}' \
  --output-dir <absolute/path>

Edit (single ref):

runcomfy run openai/gpt-image-2/edit \
  --input '{
    "prompt": "<edit instruction>",
    "images": ["https://..."]
  }' \
  --output-dir <absolute/path>

Edit (multi-ref, up to 10):

runcomfy run openai/gpt-image-2/edit \
  --input '{
    "prompt": "compose subject from image 1 into the room from image 2; match the lighting of image 2",
    "images": ["https://...subject.jpg", "https://...room.jpg"]
  }' \
  --output-dir <absolute/path>

The CLI submits, polls every 2s until terminal, then downloads any *.runcomfy.net / *.runcomfy.com URL from the result into --output-dir. Stdout is the result JSON. Stderr is progress.

For pipe-friendly usage:

runcomfy --output json run openai/gpt-image-2/text-to-image \
  --input '{"prompt":"..."}' --no-wait | jq -r .request_id

Prompting — what actually works

These are model-specific patterns that empirically improve output quality. Apply to text-to-image and edit alike.

Be explicit on subject + setting + mood. "A close-up of a matte ceramic water bottle on warm linen, soft window light, neutral background" — three concrete directives — beats "nice product photo of a bottle".

Quote embedded text exactly. Keep it short. GPT Image 2 is the strongest text-rendering model in this class, but only when you put the literal characters in quotes. Long blocks of text degrade. For multilingual text, name the script: "Japanese kana", "Cyrillic", "Arabic right-to-left".

Use compositional cues directly. "rule of thirds", "close-up", "aerial view", "centered subject", "shallow depth of field" — these have learned-meaning to the model.

Iterate one attribute at a time. When refining, change one thing per iteration (lighting OR background OR pose OR text) and keep the rest of the prompt verbatim. The model holds composition stable across iterations when only one knob moves.

Don't conflict instructions. "no text" + "the word 'AQUA+' on the label" is incoherent — the model will pick one and you don't control which.

Don't pile up styles. "ukiyo-e + watercolor + 8K + cinematic + minimalist" cancels out. Pick one or two style anchors max.

For the edit endpoint specifically:

  • State preservation goals. "keep the person's pose and face identity unchanged", "keep the brand mark and typography on the package", "keep the overall framing". The model needs to know what NOT to change.
  • Use directional language for spatial edits. "Move the headline from top-right to bottom-center", not "reposition the headline".
  • Multi-ref: number the images in the prompt — "subject from image 1, lighting and background from image 2" — and the model will route the cues correctly.

Where it shines

Use caseWhy GPT Image 2
E-commerce product photographyReliable text on labels, brand-safe lighting, consistent across SKUs
High-conversion adsHeadline + visual integration in one pass
Brand asset localizationOne source asset → many language variants of the same headline
Signage, posters, packaging mock-upsText rendering accuracy at multiple scales
UI mockups, scientific illustrationsLayout precision and label legibility

Sample prompts (verified to produce strong results)

Text-to-image — product hero:

A minimal hero product still life: a matte ceramic water bottle on warm linen,
soft window light, the word "AQUA+" in clean sans-serif on the label,
subtle rim highlights, e-commerce ready, 8K detail, neutral background

Text-to-image — multilingual signage:

A small Tokyo café storefront at dusk, warm interior glow,
the sign reads "コーヒー" in bold Japanese kana on a wooden plaque,
shallow depth of field, rule of thirds, cinematic

Edit — background swap with preservation:

Turn the background into a bright minimal white-to-soft-gray studio sweep
with gentle floor shadow; add a large headline in-image that reads
"OPEN STUDIO" in a bold clean sans-serif, high contrast, centered;
keep the main person or product, pose, and face identity unchanged

Limitations

  • Only 3 fixed sizes on text-to-image (and the same 3 + auto on edit). Extreme aspect ratios are auto-resized to the nearest supported one.
  • Prompt length ~ a few thousand tokens. Long blocks of embedded text degrade output.
  • Edit's multi-image support is "guidance from up to 10 refs", not ControlNet-style stacks. The first image is treated as the primary; the rest provide auxiliary cues.
  • Photorealism on portraits is not its strongest suit — Nano Banana Pro wins that head-to-head.

Exit codes

The runcomfy CLI uses sysexits-style codes:

codemeaning
0success
64bad CLI args
65bad input JSON / schema mismatch (e.g. size: "2048_2048" would 422)
69upstream 5xx
75retryable: timeout / 429
77not signed in or token rejected

Full reference: docs.runcomfy.com/cli/troubleshooting.

How it works

  1. The skill invokes runcomfy run openai/gpt-image-2/<endpoint> with a JSON body matching the schema above.
  2. The CLI POSTs to https://model-api.runcomfy.net/v1/models/openai/gpt-image-2/<endpoint> with the user's bearer token.
  3. The Model API returns a request_id; the CLI polls GET .../requests/<id>/status every 2 seconds.
  4. On terminal status, the CLI fetches GET .../requests/<id>/result and downloads any URL whose host ends with .runcomfy.net or .runcomfy.com into --output-dir. Other URLs are listed but not fetched.
  5. Ctrl-C while polling sends POST .../requests/<id>/cancel so you don't get billed for GPU you stopped.

What this skill is not

Not a direct OpenAI API client. Not a capability grant — depends on a working RunComfy account. Not multi-tenant.

Security & Privacy

  • Token storage: runcomfy login writes the API token to ~/.config/runcomfy/token.json with mode 0600 (owner-only read/write). Set RUNCOMFY_TOKEN env var to bypass the file entirely in CI / containers.
  • Input boundary: the user prompt is passed as a JSON string to the CLI via --input. The CLI does NOT shell-expand the prompt; it transmits the JSON body directly to the Model API over HTTPS. No shell injection surface from prompt content.
  • Third-party content: image / mask / video URLs you pass are fetched by the RunComfy model server, not by the CLI on your machine. Treat external URLs as untrusted; image-based prompt injection is a known risk for any image-edit / video-edit model.
  • Outbound endpoints: only model-api.runcomfy.net (request submission) and *.runcomfy.net / *.runcomfy.com (download whitelist for generated outputs). No telemetry, no callbacks.
  • Generated-file size cap: the CLI aborts any single download > 2 GiB to prevent disk-fill from a malicious or runaway model output.

More skills from runcomfy-com

video-inpainting
runcomfy-com
Region edits across video frames on RunComfy via the `runcomfy` CLI — remove an object that appears across many frames, clean up wires or watermarks, replace a region with matching motion. Routes across Wan 2-7 edit-video (default, prompt-driven region edits with spatial language), Lucy Edit Restyle (identity-stable region-aware restyle), and Seedream 4-0 edit-sequential (when treating the clip as a frame stack). Picks the right route based on whether the change is prose-driven,...
videocreativemedia
runcomfy-cli
runcomfy-com
Run any model on RunComfy from the command line. The `runcomfy` CLI is one binary, one auth, hundreds of model endpoints — image generation, image edit, video generation, image-to-video, lip-sync, face swap, video edit, inpainting, outpainting, extend, ControlNet, relight, upscale, LoRA training and more. Submit a request, poll for status, download the output. This skill teaches the agent how to install, authenticate, discover model schemas, invoke models, stream / poll / no-wait, script in...
creativemediaapi
image-to-video
runcomfy-com
Animate any still image on RunComfy — this skill is a smart router that matches the user's intent to the right i2v model in the RunComfy catalog. Picks HappyHorse 1.0 I2V (Arena #1, native audio, identity preservation) for general animations, Wan 2.7 with `audio_url` for custom-voiceover lip-sync, or Seedance 2.0 Pro for multi-modal animation from image + reference video + reference audio. Bundles each model's documented prompting patterns so the caller gets sharper output without burning...
creativevideomedia
flux-2-klein
runcomfy-com
Generate images with Flux 2 Klein (Black Forest Labs' distilled fast variant of Flux 2) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Flux 2 Klein's strengths (sub-second latency, multi-reference brand styling, declarative subject-first prompts), the step-count strategy (4–8 for fast iteration, ~25 for polish), the 9B vs 4B variant trade-off, and when to route to Flux 2 Pro /...
creativeimageresearch
ai-avatar-video
runcomfy-com
Create AI avatar, talking-head, and lip-sync videos on RunComfy via the `runcomfy` CLI. Routes across ByteDance OmniHuman (audio-driven full-body avatar), Wan-AI Wan 2-7 (audio-driven mouth sync via `audio_url` on a portrait), HappyHorse 1.0 (Arena #1 t2v / i2v with in-pass audio), and Seedance v2 Pro (multi-modal cinematic with reference audio + reference subject). Picks the right model for the user's actual intent — UGC voiceover, virtual presenter, dubbed product demo, lip-synced...
videocreativemedia
nano-banana-edit
runcomfy-com
Edit images with Google Nano Banana 2 (image-to-image edit endpoint) on RunComfy. Documents Nano Banana Edit's strengths (preserve subject identity, swap background, localize edits with spatial language, multi-image batch edits up to 20 inputs), the schema, and when to route to GPT Image 2 edit / Flux Kontext / Nano Banana 2 t2i instead. Calls `runcomfy run google/nano-banana-2/edit` through the local RunComfy CLI. Triggers on "nano banana edit", "edit with nano banana", "image edit nano...
creativeimageapi
wan-2-7
runcomfy-com
Generate text-to-video with Wan 2.7 (Wan-AI's flagship motion model) on RunComfy. Documents Wan 2.7's strengths (multi-reference conditioning, audio-driven lip-sync via `audio_url`, smoother transitions, prompt expansion), the duration / resolution / aspect-ratio schema, and when to route to HappyHorse 1.0 / Seedance 2.0 / Kling / LTX 2 instead. Calls `runcomfy run wan-ai/wan-2-7/text-to-video` through the local RunComfy CLI. Triggers on "wan", "wan 2.7", "wan-2-7", "wan video", or any...
creativevideomedia
lipsync
runcomfy-com
Lip-sync a face to a specific audio track on RunComfy via the `runcomfy` CLI. Routes across ByteDance OmniHuman (audio-driven full-body avatar from a portrait + audio), Sync Labs sync v2 / Pro (state-of-the-art mouth sync onto a video), Kling lipsync (audio-to- video and text-to-video with synced speech), and Creatify lipsync. The skill picks the right endpoint for the user's actual intent — portrait still + audio (avatar-style), source video + audio (mouth- swap on existing footage), or...
creativevideomedia