Polaris-1.7B-Preview-f32-GGUF

Polaris is an open-source post-training method that uses reinforcement learning (RL) scaling to refine and enhance models with advanced reasoning abilities. Our research shows that even top-tier models like Qwen3-4B can achieve significant improvements on challenging reasoning tasks when optimized with Polaris. By leveraging open-source data and academic-level resources, Polaris pushes the capabilities of open-recipe reasoning models to unprecedented heights.

Model Files

File Name	Quant Type	Size
Polaris-1.7B-Preview.BF16.gguf	BF16	3.45 GB
Polaris-1.7B-Preview.F16.gguf	F16	3.45 GB
Polaris-1.7B-Preview.F32.gguf	F32	6.89 GB
Polaris-1.7B-Preview.Q2_K.gguf	Q2_K	778 MB
Polaris-1.7B-Preview.Q3_K_L.gguf	Q3_K_L	1 GB
Polaris-1.7B-Preview.Q3_K_M.gguf	Q3_K_M	940 MB
Polaris-1.7B-Preview.Q3_K_S.gguf	Q3_K_S	867 MB
Polaris-1.7B-Preview.Q4_K_M.gguf	Q4_K_M	1.11 GB
Polaris-1.7B-Preview.Q4_K_S.gguf	Q4_K_S	1.06 GB
Polaris-1.7B-Preview.Q5_K_M.gguf	Q5_K_M	1.26 GB
Polaris-1.7B-Preview.Q5_K_S.gguf	Q5_K_S	1.23 GB
Polaris-1.7B-Preview.Q6_K.gguf	Q6_K	1.42 GB
Polaris-1.7B-Preview.Q8_0.gguf	Q8_0	1.83 GB

Quants Usage

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):