gpt-image-2

Name: gpt-image-2
Author: prime-skills

31stars9forks

How to install gpt-image-2?

prime-skills/runcomfy-agent-skills/gpt-image-2

$npx -y skills add prime-skills/runcomfy-agent-skills --skill gpt-image-2

Installs into the current project.

Summary

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

Files · 1

View on GitHub

SKILL.md

1	# GPT Image 2 — Pro Pack on RunComfy
2
3	[runcomfy.com](https://www.runcomfy.com/?utm_source=skills.sh&utm_medium=skill&utm_campaign=gpt-image-2) · [Text-to-image](https://www.runcomfy.com/models/openai/gpt-image-2/text-to-image?utm_source=skills.sh&utm_medium=skill&utm_campaign=gpt-image-2) · [Edit](https://www.runcomfy.com/models/openai/gpt-image-2/edit?utm_source=skills.sh&utm_medium=skill&utm_campaign=gpt-image-2) · [GitHub](https://github.com/agentspace-so/runcomfy-skills/tree/main/gpt-image-2)
4
5	OpenAI GPT Image 2 (ChatGPT Images 2.0) hosted on the RunComfy Model API — no OpenAI key, async REST.
6
7	```bash
8	npx skills add agentspace-so/runcomfy-skills --skill gpt-image-2 -g
9	```
10
11	## When to pick this model (vs siblings)
12
13	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.
14
15	\| You want \| Use \|
16	\|---\|---\|
17	\| Embedded text, logos, signage, multilingual typography \| GPT Image 2 \|
18	\| Brand-safe, e-commerce / ad / UI mockup imagery \| GPT Image 2 \|
19	\| Iterative refinement that holds composition stable \| GPT Image 2 \|
20	\| Heavy stylization, painterly look \| Flux 2 \|
21	\| Hyperrealistic portrait \| Nano Banana Pro \|
22	\| Cinematic / aesthetic-first hero shots \| Seedream 5 \|
23
24	If the user explicitly asked for GPT Image 2 / ChatGPT Image 2 / Image 2, route here regardless — don't second-guess the model choice.
25
26	## Prerequisites
27
28	1. RunComfy CLI — `npm i -g @runcomfy/cli`
29	2. RunComfy account — `runcomfy login` opens a browser device-code flow.
30	3. CI / containers — set `RUNCOMFY_TOKEN=<token>` instead of `runcomfy login`.
31
32	## Endpoints + input schema
33
34	Two endpoints, same model.
35
36	### `openai/gpt-image-2/text-to-image`
37
38	\| Field \| Type \| Required \| Default \| Notes \|
39	\|---\|---\|---\|---\|---\|
40	\| `prompt` \| string \| yes \| — \| The positive prompt \|
41	\| `size` \| enum \| no \| `1024_1024` \| `1024_1024` (1:1), `1024_1536` (2:3 portrait), `1536_1024` (3:2 landscape) — only these three \|
42
43	### `openai/gpt-image-2/edit`
44
45	\| Field \| Type \| Required \| Default \| Notes \|
46	\|---\|---\|---\|---\|---\|
47	\| `prompt` \| string \| yes \| — \| Natural-language edit instruction \|
48	\| `images` \| string[] \| yes \| — \| Up to 10 reference image URLs (publicly fetchable HTTPS) \|
49	\| `size` \| enum \| no \| `auto` \| `auto` (preserve input ratio), or one of the three fixed sizes above \|
50
51	`size=auto` on edit preserves the input aspect ratio — strongly recommended unless the edit explicitly changes framing.
52
53	## How to invoke
54
55	Text-to-image:
56
57	```bash
58	runcomfy run openai/gpt-image-2/text-to-image \
59	--input '{"prompt": "<user prompt>", "size": "1024_1536"}' \
60	--output-dir <absolute/path>
61	```
62
63	Edit (single ref):
64
65	```bash
66	runcomfy run openai/gpt-image-2/edit \
67	--input '{
68	"prompt": "<edit instruction>",
69	"images": ["https://..."]
70	}' \
71	--output-dir <absolute/path>
72	```
73
74	Edit (multi-ref, up to 10):
75
76	```bash
77	runcomfy run openai/gpt-image-2/edit \
78	--input '{
79	"prompt": "compose subject from image 1 into the room from image 2; match the lighting of image 2",
80	"images": ["https://...subject.jpg", "https://...room.jpg"]
81	}' \
82	--output-dir <absolute/path>
83	```
84
85	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.
86
87	For pipe-friendly usage:
88
89	```bash
90	runcomfy --output json run openai/gpt-image-2/text-to-image \
91	--input '{"prompt":"..."}' --no-wait \| jq -r .request_id
92	```
93
94	## Prompting — what actually works
95
96	These are model-specific patterns that empirically improve output quality. Apply to text-to-image and edit alike.
97
98	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".
99
100	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"

People who liked this also viewed

6 picks

Type

gpt-image-2

31stars9forks

View on GitHub

How to install gpt-image-2?

prime-skills/runcomfy-agent-skills/gpt-image-2

$npx -y skills add prime-skills/runcomfy-agent-skills --skill gpt-image-2

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# GPT Image 2 — Pro Pack on RunComfy
2
3	[runcomfy.com](https://www.runcomfy.com/?utm_source=skills.sh&utm_medium=skill&utm_campaign=gpt-image-2) · [Text-to-image](https://www.runcomfy.com/models/openai/gpt-image-2/text-to-image?utm_source=skills.sh&utm_medium=skill&utm_campaign=gpt-image-2) · [Edit](https://www.runcomfy.com/models/openai/gpt-image-2/edit?utm_source=skills.sh&utm_medium=skill&utm_campaign=gpt-image-2) · [GitHub](https://github.com/agentspace-so/runcomfy-skills/tree/main/gpt-image-2)
4
5	OpenAI GPT Image 2 (ChatGPT Images 2.0) hosted on the RunComfy Model API — no OpenAI key, async REST.
6
7	```bash
8	npx skills add agentspace-so/runcomfy-skills --skill gpt-image-2 -g
9	```
10
11	## When to pick this model (vs siblings)
12
13	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.
14
15	\| You want \| Use \|
16	\|---\|---\|
17	\| Embedded text, logos, signage, multilingual typography \| GPT Image 2 \|
18	\| Brand-safe, e-commerce / ad / UI mockup imagery \| GPT Image 2 \|
19	\| Iterative refinement that holds composition stable \| GPT Image 2 \|
20	\| Heavy stylization, painterly look \| Flux 2 \|
21	\| Hyperrealistic portrait \| Nano Banana Pro \|
22	\| Cinematic / aesthetic-first hero shots \| Seedream 5 \|
23
24	If the user explicitly asked for GPT Image 2 / ChatGPT Image 2 / Image 2, route here regardless — don't second-guess the model choice.
25
26	## Prerequisites
27
28	1. RunComfy CLI — `npm i -g @runcomfy/cli`
29	2. RunComfy account — `runcomfy login` opens a browser device-code flow.
30	3. CI / containers — set `RUNCOMFY_TOKEN=<token>` instead of `runcomfy login`.
31
32	## Endpoints + input schema
33
34	Two endpoints, same model.
35
36	### `openai/gpt-image-2/text-to-image`
37
38	\| Field \| Type \| Required \| Default \| Notes \|
39	\|---\|---\|---\|---\|---\|
40	\| `prompt` \| string \| yes \| — \| The positive prompt \|
41	\| `size` \| enum \| no \| `1024_1024` \| `1024_1024` (1:1), `1024_1536` (2:3 portrait), `1536_1024` (3:2 landscape) — only these three \|
42
43	### `openai/gpt-image-2/edit`
44
45	\| Field \| Type \| Required \| Default \| Notes \|
46	\|---\|---\|---\|---\|---\|
47	\| `prompt` \| string \| yes \| — \| Natural-language edit instruction \|
48	\| `images` \| string[] \| yes \| — \| Up to 10 reference image URLs (publicly fetchable HTTPS) \|
49	\| `size` \| enum \| no \| `auto` \| `auto` (preserve input ratio), or one of the three fixed sizes above \|
50
51	`size=auto` on edit preserves the input aspect ratio — strongly recommended unless the edit explicitly changes framing.
52
53	## How to invoke
54
55	Text-to-image:
56
57	```bash
58	runcomfy run openai/gpt-image-2/text-to-image \
59	--input '{"prompt": "<user prompt>", "size": "1024_1536"}' \
60	--output-dir <absolute/path>
61	```
62
63	Edit (single ref):
64
65	```bash
66	runcomfy run openai/gpt-image-2/edit \
67	--input '{
68	"prompt": "<edit instruction>",
69	"images": ["https://..."]
70	}' \
71	--output-dir <absolute/path>
72	```
73
74	Edit (multi-ref, up to 10):
75
76	```bash
77	runcomfy run openai/gpt-image-2/edit \
78	--input '{
79	"prompt": "compose subject from image 1 into the room from image 2; match the lighting of image 2",
80	"images": ["https://...subject.jpg", "https://...room.jpg"]
81	}' \
82	--output-dir <absolute/path>
83	```
84
85	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.
86
87	For pipe-friendly usage:
88
89	```bash
90	runcomfy --output json run openai/gpt-image-2/text-to-image \
91	--input '{"prompt":"..."}' --no-wait \| jq -r .request_id
92	```
93
94	## Prompting — what actually works
95
96	These are model-specific patterns that empirically improve output quality. Apply to text-to-image and edit alike.
97
98	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".
99
100	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"

People who liked this also viewed

6 picks

Type