Gemini-3.1-Pro-Qwen3-14B.gguf

This is a gguf quantized version of Cannae-AI/Gemini-3.1-Pro-Qwen3-14B

Gemini-3.1-Pro-Qwen3-14B is a fine-tuned model designed for expert level reasoning tasks.This fine-tune leverages the advanced capabilities of Gemini 3.1 Pro to significantly enhance the model’s analytical depth, logical coherence, and ability to synthesize conflicting information. It is ideal for research and applications requiring advanced analytical and reasoning capabilities.

Data and finetune Description:

Finetune method: SFT
Hardware used: Nvidia L4
Data Type: Synthetic, high-complexity reasoning corpus
Prompting Agent: Gemini 3.1 Flash
Solving Agent: Gemini 3.1 Pro
Content: Expert-level problems requiring multi-step reasoning, derivations, and synthesis of conflicting information
Purpose: To push the limits of modern reasoning models in challenging domains
Size & Scope: Designed for extreme difficulty scenarios; includes a diverse set of complex problems across logic, mathematics, and domain-specific reasoning.

Intended Use:

Advanced reasoning and problem-solving applications
Research in multi-step logical reasoning and AI cognition

Limitations:

The model is trained on synthetic, high-complexity reasoning tasks; real-world performance may vary.
Specialized focus on reasoning may reduce performance in casual conversation or general NLP tasks.

Performance Notes:

Excels in multi-step reasoning, derivation, and problem synthesis.
Performance has been validated against challenging synthetic scenarios designed to stress-test reasoning capabilities.

Contact & Citation:

For questions, collaboration, or citation, please contact the Cannae-AI research team responsible for Cannae-AI/Gemini-3.1-Pro-Qwen3-14B.

Example usage:

For text only LLMs: llama-cli -hf CannaeAI/Gemini-3.1-Pro-Qwen3-14B.gguf --jinja
For multimodal models: llama-mtmd-cli -hf CannaeAI/Gemini-3.1-Pro-Qwen3-14B.gguf --jinja