lookarooka/lookarooka-brain-v1

VISIONConcurrency Cost:1Model Size:5.1BQuant:BF16Ctx Length:32kTool Calling:SupportedPublished:Jun 23, 2026Architecture:Transformer Cold

The lookarooka/lookarooka-brain-v1 is a 5.1 billion parameter language model, fine-tuned and converted to GGUF format using Unsloth. This model is available in both text-only and multimodal (vision-capable) variants, supporting a 32768 token context length. It is designed for efficient deployment and inference, particularly for applications requiring either pure text generation or combined text and image understanding.

Loading preview...

lookarooka-brain-v1 Overview

lookarooka/lookarooka-brain-v1 is a 5.1 billion parameter model, fine-tuned and optimized for GGUF format using the Unsloth framework, which facilitated 2x faster training. This model offers both text-only and multimodal capabilities, supporting a substantial 32768 token context length.

Key Capabilities

  • Efficient GGUF Format: Provided in GGUF for optimized local inference.
  • Multimodal Support: Includes a vision-capable variant (gemma-4-e2b-it.BF16-mmproj.gguf) for processing both text and images.
  • Unsloth Optimization: Benefits from Unsloth's training efficiencies, enabling faster development cycles.

Deployment Notes

  • Ollama Compatibility: Specific instructions are provided for creating a unified bf16 model for Ollama, as it currently does not support separate mmproj files for vision models.
  • Command-Line Interface: Examples are given for llama-cli for text-only models and llama-mtmd-cli for multimodal models, utilizing the --jinja flag.

Finetunes
1 models