BAGEL

ByteDanceOpenApache 2.0

BAGEL is an AI image model from ByteDance. On LMArena’s Text-to-image leaderboard it ranks 84 of 84 with 897±6 from 12,361 votes. On LMArena’s Image editing leaderboard it ranks 62 of 63 with 1026±6 from 13,488 votes. License: Apache 2.0.

Key figures

LMArena · Text-to-image897±6#84 / 84 · Votes 12,361
LMArena · Image editing1026±6#62 / 63 · Votes 13,488
AA Elo (Text-to-image)594±35Bagel
AA Elo (Image editing)789±25Bagel
Developer
ByteDance
License
Apache 2.0
Released (per AA)
May 2025

Who it is for (from the data)

  • When quality is not the top priority: #84 of 84 on Text-to-image, in the bottom half; if quality matters most, check higher-ranked models first.
  • Generating and editing images: it is ranked on both boards.
  • Self-hosting or on-premise use: LMArena lists its license as Apache 2.0.
  • Existing ByteDance users: this vendor has 8 models on Text-to-image.

Neighbours on Text-to-image

#ModelScoreVotesAA EloLicenseΔ
78GLM-Image Z.ai10124,856885MIT+115
79Gemini 2.0 Flash Preview Image Generation Google976257,205–Proprietary+79
80FLUX.1 [dev] (FP8) Black Forest Labs96949,221826Open+72
81DALL-E 3 OpenAI968238,732–Proprietary+71
82FLUX.1 Kontext [dev] Black Forest Labs941216,216–flux-1-dev-non-commercial-license+44
83?Stable Diffusion 3.5 Large 93823,396–Open+41
84BAGEL ByteDance89712,361594Apache 2.0

Alternatives · More from ByteDance

#ModelScoreVotesAA EloLicense
16Seedream 5.0 Pro ByteDance1250112,5851101Proprietary
38Seedream 4.5 ByteDance1148323,6751027Proprietary
41Seedream 5.0 Lite ByteDance1141148,3711019Proprietary
42Seedream 4.0 (2K) ByteDance113912,563–Proprietary
49Seedream 4.0 (fal) ByteDance111611,840–Proprietary

Alternatives · Top open-weight models

#ModelScoreVotesAA EloLicense
21Qwen-Image-2.1 Alibaba122719,3901043qwen-research
22Ideogram 4.0 (Quality) Ideogram119980,6461025Ideogram Open Model
29Cosmos3-Super-Text2Image (Agentic) NVIDIA116510,751993OpenMDW-1.1
35Cosmos3-Super-Text2Image NVIDIA115118,368986OpenMDW-1.1
36HunyuanImage 3.0 Tencent1151180,470932tencent-hunyuan-community

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 BAGEL rank among AI image models?

Text-to-image rank 84 (897); Image editing rank 62 (1026) (LMArena, data as of Nov 7, 2026).

Is BAGEL open-weight?

LMArena lists its license as Apache 2.0 (not proprietary); check the license text for commercial-use terms.

Can BAGEL edit images?

Yes. BAGEL is #62 of 63 on the LMArena image-edit board (1026, 13,488 votes).

Who makes BAGEL?

According to LMArena, BAGEL is made by ByteDance.

How reliable is BAGEL's score?

The scores are based on 12,361 votes on Text-to-image (95% CI ±6); 13,488 votes on Image editing (95% CI ±6). More votes and narrower intervals mean a more stable rank.

How far is BAGEL from #1 on Text-to-image?

#1 on Text-to-image is GPT-Image-2.5 Sunburst (1429); BAGEL scores 897, 532 points behind.

Sources

LMArena Text-to-image data published on Nov 6, 2026.

LMArena Image editing data published on Nov 6, 2026.

17NAS syncs the upstream leaderboards daily; this page's data was synced on Nov 7, 2026.

Price, resolution and speed have no reliable public data yet and are omitted; all values come from the snapshot, blanks show –.