全站搜索

AI 模型 84759

• Phantom_Wan_14B_FusionX-GGUF↗

  QuantStack

• Hunyuan-A13B-Instruct-hf-WIP-GGUF↗

  qwp4w3hyb

• Muse-Spark-Open-DFlash2↗

  RadixArk

• ai-text-detector-qwen3-0.6b-variable↗

  rasbt

• ai-human-lab_-_EEVE-Korean-10.8B-instruct-v4-sft-gguf↗

  RichardErkhov

• m-a-p_-_OpenCodeInterpreter-DS-6.7B-gguf↗

  RichardErkhov

• rizzo-flow↗

  rizzoaiacademy

  These are LoRA fine-tunes of XHToken/Spark-X2.5-4B (recommended) and XHToken/Spark-X2.5-1.7B, for Rizzo Flow. Rizzo Flow turns an unstructured state into typed decisions (`boolean`, `choice`, `score`) with probabilities, and it **generates no text**. The model runs one forward pass per question. The answer is the softmax over the logits of the answer letters (`A`, `B`, …) at the last prompt position. Training works the same way. There is no text target: the loss is a soft cross-entropy between the target distribution and the softmax over the answer letters. Only the LoRA adapters are trained. These GGUF files are what Rizzo Flow downloads by default. It pins this repository by commit and checks the sha256 of every file:

• saina-helm-2-omni-0.8b↗

  run-saina

• ReactionT5v2-yield↗

  sagawa

• Gemma-4-12B-it-heretic-GGUF↗

  SC117