Top 10 Most Popular AI Models on Replicate (February 2026)

GFPGAN

🔥 108.2M runs Image Restoration

Practical face restoration algorithm for old photos or AI-generated faces. Developed by Tencent ARC Lab. The undisputed leader on Replicate.

multilingual-e5-large

58.9M runs Text Embedding

A multi-language text embedding model. Essential for semantic search and RAG applications across 100+ languages.

Nano Banana Pro

12.8M runs Image Generation Official

Google's state of the art image generation and editing model (Gemini 3 Pro Image). SOTA text rendering and precise editing controls.

Z-Image Turbo

20.5M runs Text-to-Image Official

Super fast text-to-image model of 6B parameters developed by Tongyi-MAI. Optimized for production use with sub-second generation.

Seedream 4.5

3M runs Image Generation Official

Upgraded ByteDance image model with stronger spatial understanding and world knowledge. Strong competitor to Google's models.

P-Image Edit

12.4M runs Image Editing Official

A sub 1 second $0.01 multi-image editing model built for production use cases. Cost-effective batch processing.

FLUX.2 [klein] 4B

2.7M runs Image Generation Official

Very fast image generation and editing model from Black Forest Labs. 4 steps distilled, sub-second inference for production.

GPT-Image 1.5

2.4M runs Image Generation Official

OpenAI's latest image generation model with better instruction following and adherence to prompts. Released February 2026.

FLUX.2 [klein] 9B

41.8K runs Image Generation Official

4 step distilled version of FLUX.2 [klein]. A foundation model for maximum flexibility and control from Black Forest Labs.

#10

Gemini 3 Flash

134.9K runs LLM / Vision Official

Google's most intelligent model built for speed with frontier intelligence, superior search, and grounding. The only LLM in the top 10.

Image Generation (60%): 6 out of 10 models - Nano Banana Pro, Z-Image Turbo, Seedream 4.5, FLUX.2 variants, GPT-Image 1.5
Image Editing (20%): P-Image Edit - production-focused batch editing
Image Restoration (10%): GFPGAN - the runaway leader with 108M runs
Text Embedding (10%): multilingual-e5-large - essential infrastructure
LLM/Vision (10%): Gemini 3 Flash - only language model in top 10