Pinkstack/Superthoughts-lite-v2-MOE-Llama3.2-bf16

BFloat16 version of Superthoughts lite v2 MoE. 3.91B parameters, 2 experts active, 4 in total. (recommended, float32 is heavy.)

[GGUF !! Full precision !! BF16]

INFORMATION

This is the non-experimental version of Superthoughts Lite v2. Offering better accuracy at all tasks, better performance and less looping while generating responses.

We trained it by first creating a base model for all the experts, which was fine-tuned using GRPO techniques using Unsloth on top of meta-llama/Llama-3.2-1B-Instruct. After making the base model, we trained each potential expert using SFT. After doing SFT, we did GRPO again. in total there are 4 experts:

• Chat reasoning expert,
• Math reasoning expert,
• Code reasoning expert,
• Science reasoning expert.

By doing this, we obtained a powerful, lite reasoning model that is very usable for its size.

This model is a direct replacement of Pinkstack/Superthoughts-lite-v1. Pinkstack/Superthoughts-lite-v1 was not able to generate code, and had very poor text performance. V2 is much more usable.

You should use this system prompt:

Thinking: enabled.

Follow this format strictly:
<think>
Write your step-by-step reasoning here.
Break down the problem into smaller parts.
Solve each part systematically.
Check your work and verify the answer makes sense.
</think>
[Your final answer after thinking].

Model information

The model can generate up to 16,380 tokens and has a context size of 131072 It has been fine tuned to generated thinking data in-between <think> xml tags. note that it may still have some slight looping but they are rare.

LIMITATIONS

While some safety alignment was done by us, it was very minimal. Thus, the model can be uncensored at times. In addition, users and providers alike should be aware that all large language models (LLM's), including this one can hallucinate and output false information. Always double check responses.

By using this model, you agree to the LLAMA 3.2 COMMUNITY LICENSE.