TLDR AI · ENGINEERING & RESEARCH
PPLX-DECIDER-V1-27B (HUGGING FACE REPO)
Summary from the TLDR AI newsletter
pplx-decider-v1-27b is a decision model fine-tuned from Qwen3.8-27B. Its benchmark results are competitive with Jev, beating it in several tests.