Imagen 4 (generate-001)
GoogleProprietary
Imagen 4 (generate-001) is an AI image model from Google. On LMArena’s Text-to-image leaderboard it ranks 43 of 82 with 1128±3 from 539,657 votes. It is not on LMArena’s Image editing leaderboard yet. License: Proprietary.
Key figures
- Developer
- License
- Proprietary
Not yet available: Released (per AA), LMArena · Image editing, AA Elo
Who it is for (from the data)
- When quality is not the top priority: #43 of 82 on Text-to-image, in the bottom half; if quality matters most, check higher-ranked models first.
- Mainly text-to-image: no image-edit results yet.
- Use via Google's service or API: LMArena lists it as proprietary.
- Existing Google users: this vendor has 10 models on Text-to-image.
Neighbours on Text-to-image
| # | Model | Score | Votes | AA Elo | License | Δ |
|---|---|---|---|---|---|---|
| 40 |  Seedream 4.0 (2K) ByteDance | 1139 | 12,563 | – | Proprietary | +11 |
| 41 |  Wan 2.6 Alibaba | 1138 | 247,795 | – | Proprietary | +10 |
| 42 |  Recraft V4.1 Pro Recraft | 1131 | 2,796 | – | Proprietary | +3 |
| 43 | Imagen 4 (generate-001) Google | 1128 | 539,657 | – | Proprietary | |
| 44 |  Krea 2 Medium Krea | 1128 | 69,957 | 1012 | Proprietary | +0 |
| 45 | Qwen-Image (2512) Alibaba | 1125 | 98,272 | – | Apache 2.0 | -3 |
| 46 | Wan 2.5 Preview Alibaba | 1117 | 290,257 | – | Proprietary | -11 |
Alternatives · More from Google
| # | Model | Score | Votes | AA Elo | License |
|---|---|---|---|---|---|
| 5 | Gemini Nano Banana 2.1 Google | 1325 | 5,290 | 1176 | Proprietary |
| 11 | Gemini 3.1 Flash Image (nano-banana-2, Web Search) Google | 1257 | 73,278 | 1140 | Proprietary |
| 15 | Gemini 3.1 Flash-Lite Image (nano-banana-2-lite) Google | 1249 | 17,813 | 1107 | Proprietary |
| 16 | Gemini 3 Pro Image (nano-banana-pro, 2K) Google | 1246 | 188,288 | 1118 | Proprietary |
| 18 | Gemini 3 Pro Image Preview (nano-banana-pro) Google | 1232 | 86,470 | – | Proprietary |
Alternatives · Top open-weight models
| # | Model | Score | Votes | AA Elo | License |
|---|---|---|---|---|---|
| 19 | Qwen-Image-2.1 Alibaba | 1225 | 19,604 | 1042 | qwen-research |
| 20 |  Ideogram 4.0 (Quality) Ideogram | 1200 | 72,865 | 1026 | Ideogram Open Model |
| 27 |  Cosmos3-Super-Text2Image (Agentic) NVIDIA | 1164 | 10,803 | 993 | OpenMDW-1.1 |
| 33 |  HunyuanImage 3.0 Tencent | 1151 | 180,498 | 932 | tencent-hunyuan-community |
| 34 | Cosmos3-Super-Text2Image NVIDIA | 1151 | 18,480 | 985 | OpenMDW-1.1 |
How to read the scores
- LMArena score: people blind-pick the better of two anonymous images for the same prompt, and the votes are fitted into an Elo-style rating; "±" is the 95% confidence interval.
- Text-to-image and image edit are separate boards, so scores are not comparable across them; AA Elo comes from Artificial Analysis on a different scale.
- If the gap between two models is smaller than their combined intervals, the order is not statistically reliable; treat them as the same tier. More votes mean narrower intervals.
FAQ
Where does Imagen 4 (generate-001) rank among AI image models?
Text-to-image rank 43 (1128) (LMArena, data as of Oct 27, 2026).
Is Imagen 4 (generate-001) open-weight?
No. LMArena lists it as Proprietary; it is available only through Google’s products or API.
Can Imagen 4 (generate-001) edit images?
The LMArena image-edit board has no result for Imagen 4 (generate-001) yet, so we cannot confirm its editing ability from the data.
Who makes Imagen 4 (generate-001)?
According to LMArena, Imagen 4 (generate-001) is made by Google.
How reliable is Imagen 4 (generate-001)'s score?
The scores are based on 539,657 votes on Text-to-image (95% CI ±3). More votes and narrower intervals mean a more stable rank.
How far is Imagen 4 (generate-001) from #1 on Text-to-image?
#1 on Text-to-image is GPT-Image-2.5 Sunburst (1427); Imagen 4 (generate-001) scores 1128, 299 points behind.
Sources
LMArena Text-to-image data published on Oct 26, 2026.
LMArena Image editing data published on Oct 26, 2026.
17NAS syncs the upstream leaderboards daily; this page's data was synced on Oct 27, 2026.
Price, resolution and speed have no reliable public data yet and are omitted; all values come from the snapshot, blanks show –.