AI Model Catalogue

Granite-4.0-H-Small is a 32B parameter long-context instruct model finetuned from Granite-4.0-H-Small-Base using a combination of open source instruction datasets with permissive license and internally collected synthetic datasets.

90's Photographs in the style of Bjorks promo material

Advanced 20B open-weight reasoning models to customize for any use case and run anywhere.

Grok 4 is xAI’s most advanced reasoning model. Excels at logical thinking and in-depth analysis. Ideal for insightful discussions and complex problem-solving.

Highest-quality text-to-speech with <200ms latency, emotion control, and 15-language support

Remove backgrounds with real alpha transparency using Nano Banana 2. Triangulation matting produces clean edges, proper semi-transparency, and accurate colors — superior to traditional background removal. Optionally specify what to isolate.

Open-source Phi-3-mini-4k-instruct model from microsoft — available for download and self-hosting on Hugging Face.

Ultralytics YOLO11n object detection model with 2.6M parameters. Achieves 39.5 mAP50-95 on COCO dataset. Optimized for real-time inference with 1.55 ms speed on T4 GPU..

This is an optimised version of the hidream-l1 model using the pruna ai optimisation toolkit!

Ultralytics YOLOv8s worldv2 Real-Time Open-Vocabulary Object Detection model with 12.7M parameters. Achieves 37.7 mAP50-95 on COCO dataset. Optimized for real-time inference

Beta/RFC version of https://replicate.com/nelsonjchen/op-replay-clipper

The smartest diffusion language model up to ~800+ tps

This is Platmoji 2, trained more to mimic emojis in an extremely similar way. (Realism in emojis go to Platmoji 1)

Ultralytics YOLOE-L Real-Time Seeing Anything model with 26.2M parameters. Achieves 52.6 mAP50-95 on COCO dataset. Optimized for real-time inference with 6.2 ms speed on T4 GPU..

This is a version of the MoE Gemma 4 26B optimised by Pruna AI.

Google's fast, expressive text-to-speech model with 30 voices and 70+ language support

Arise V2

Inspired by 90s streetwear, graffiti culture, and macabre aesthetics

Object Recognition model

Temporary publish target for bgogo deployment validation.

This is a version of Qwen 3.5 27B optimised by Pruna AI.

A 4B-parameter safety and content moderation model that classifies user prompts and assistant responses as Safe, Unsafe, or Controversial with fine-grained category labels and refusal detection. Supports 119 languages.

Universal Feed-Forward Metric 3D Reconstruction