← All genres
IMAGE
Image
Text-to-image and image editing. Split between artistic quality and open weights.
11 models tracked · 31 releases on record · last movement 2w from Unlimited-OCR
Where the genre stands
Ordered by most recently shipped
ModelVersion
- Unlimited-OCRBaiduCNOpen Weights3.3B MIT-licensed open-weight vision-language OCR model for dense document, table, and formula recognition.2
- Qwen-ImageAlibabaCNOpen Weights7.1B open-weight text-to-image and image editing foundation model from the Qwen team.2.1T
- FLUXBlack Forest LabsEUOpen WeightsThe open-weight image default, now expanding past images entirely.2T
- MidjourneyMidjourneyGlobalStill the reference point for artistic style over literal accuracy.8.5
- IdeogramIdeogramUSImage model whose distinguishing trait is that it renders text correctly.4.5
- MAI ImageMicrosoftUSMicrosoft's in-house image model, aimed at design-ready output.2.6
- SeedreamByteDanceCNRegion-precise image editing that changes one element and leaves the frame.5 Pro
- ReveReveCNPrompt-driven image editing with strong layout and text handling.2.1
- Nano BananaGoogleCNGoogle's image model — shipped under two names, one on each platform.2 Lite
- PaddleOCR-VLBaidu PaddlePaddleCNOpen WeightsCompact 0.96B Apache-2.0 vision-language document parser for multilingual text, tables, formulas, and charts.1.6
- RecraftRecraftEUState-of-the-art graphic design, branding, and typography generative foundation model.V4.1
Everything that shipped
- 2wIMAGEUnlimited-OCRBaiduCNNEW DROPBaidu releases Unlimited-OCR 2 open weights under MIT license on Hugging Face.2
- 1moIMAGEQwen-ImageAlibabaCNQwen-Image 2.1 Turbo lands on Hugging Face for few-step image synthesis.2.1T
- 1moIMAGEFLUXBlack Forest LabsEUFLUX.2 Turbo drops latency to 8 steps with complete typography prompt adherence.2T
- 1moIMAGEMidjourneyMidjourneyGlobalMidjourney V8.5 introduces direct web-canvas camera re-framing and 4K photorealism.8.5
- 1moIMAGEIdeogramIdeogramUSIdeogram 4.5 is listed on fal.ai.4.5
- 1moIMAGEQwen-ImageAlibabaCNAlibaba releases Qwen-Image 2.1 open weights on Hugging Face.2.1
- 3moIMAGEMAI ImageMicrosoftUSMAI-Image-2.6 launches at No. 2 on Arena’s text-to-image leaderboard.2.6
- 3moIMAGESeedreamByteDanceCNSeedream 5.0 Pro adds grounded, layer-separated editing.5 Pro
- 3moIMAGEMAI ImageMicrosoftUSMAI-Image-2.5-Pro introduced alongside MAI-Voice-2-Flash.2.5P
- 3moIMAGEFLUXBlack Forest LabsEUFLUX 3 enters early access — video, audio and robot actions from one model.3
- 4moIMAGEReveReveCNReve 2.1 edits from text prompts with layout intelligence.2.1
- 4moIMAGESeedreamByteDanceCNSeedream 5.0 Lite editing endpoint.5
- 4moIMAGEIdeogramIdeogramUSIdeogram V4.0 sharpens text rendering and fine detail.4.0
- 4moIMAGENano BananaGoogleCNNano Banana 2 Lite, also listed as Gemini 3.1 Flash Lite Image.2 Lite
- 4moIMAGEUnlimited-OCRBaiduCNBaidu open-sources Unlimited-OCR with 3.34B MIT-licensed weights on Hugging Face.1
- 5moIMAGEMAI ImageMicrosoftUSMAI-Image-2.5 and 2.5-Flash launch; 2.5 ranks No. 2 on Arena’s image-edit leaderboard.2.5
- 5moIMAGEPaddleOCR-VLBaidu PaddlePaddleCNPaddleOCR-VL 1.6 upgrades compact 0.96B multilingual document and table parsing on Hugging Face.1.6
- 6moIMAGERecraftRecraftEURecraft V4.1 brings quieter photorealism and new illustration styles.V4.1
- 6moIMAGEMidjourneyMidjourneyGlobalV8.1 is the fastest Midjourney yet — 4–5× on standard jobs.V8.1
- 6moIMAGEIdeogramIdeogramUSV3 with style training from reference images.3.0
- 7moIMAGEMAI ImageMicrosoftUSMAI-Image-2-Efficient: flagship quality at nearly 41% lower cost.2E
- 7moIMAGEMAI ImageMicrosoftUSMAI-Image-2 targets natural light, accurate skin tones and lived-in scenes.2
- 8moIMAGENano BananaGoogleCNNano Banana 2, Google's image generation and editing model.2
- 8moIMAGERecraftRecraftEURecraft V4, a ground-up rebuild focused on design taste.V4
- 9moIMAGEPaddleOCR-VLBaidu PaddlePaddleCNPaddleOCR-VL 1.5 ships Apache-2.0 vision-language OCR checkpoints on Hugging Face.1.5
- 9moIMAGEMidjourneyMidjourneyGlobalV8 resets the base model.V8
- 9moIMAGEFLUXBlack Forest LabsEUFLUX klein targets local and enterprise inference.klein
- 11moIMAGEFLUXBlack Forest LabsEUFLUX.2 replaces FLUX.1 as the flagship.2 Pro
- 1yIMAGEMAI ImageMicrosoftUSMAI-Image-1 debuts in the top 10 on LMArena.1
- 1yIMAGEMidjourneyMidjourneyGlobalV7 introduces draft mode and personalisation by default.V7
- 2yIMAGERecraftRecraftEURecraft V3 sets a new image-generation standard, led by text rendering.V3