jeb
4 TagsUpdated Apache-2.0by frontier-infra
Decision models by Jason Brashear and AINode: LoRAs merged into Qwen3.5 and Qwen3.8, published as GGUF. Jeb reads the option labels' logits after AINode's own decision prompt, with a fitted temperature per question type; Ollaya runs the authors' files on llama.cpp.
decisionfine-tunedgguf4b9b27b
Tags
| Name |
|---|
| jeblatestSame as jeb:9b.9.8 GB · 4096 ctx · English |
| jeb:27bJebadiah 27B (Qwen3.8), Q4_K_M GGUF, 17 GB: the largest Jeb, for a 24 GB GPU.16.8 GB · 4096 ctx · English |
| jeb:4bJebadiah 4B v2, Q8_0 GGUF: the smallest and fastest Jeb.4.6 GB · 4096 ctx · English |
| jeb:9bJebadiah 9B v2, Q8_0 GGUF: the authors' recommended local model.9.8 GB · 4096 ctx · English |
Tags without a suffix load fp16 on a CUDA GPU and fp32 on CPU; add -fp16 or -fp32 to pin one precision.