GGUF quantizations of GestaltLabs/Ornstein3.8-27B: a Qwen 3.8 27B dense vision-language fine-tune with interleaved linear and full attention (Gated DeltaNet). The LoRA was trained on Fireworks AI and merged into the Qwen3.8-27B language stack. Language GGUFs are qwen35 text-path only (866 tensors). For llama.cpp vision, pass —mmproj mmproj-Ornstein3.8-27B-BF16.gguf. For Transformers or vLLM multimodal inference, use the base safetensors. This checkpoint injects Ornstein thinking into Qwen3.8-27B. It is an early merge, not a finished quality release. Planned quality work uses RL environments and energy-based fine-tuning. Qwen3.8-27B achieves an estimated 97.0% accuracy on the full GSM8K benchmark when running in standard unquantized precision (BF16/FP8). Single greedy BF16 run on a Fireworks dedicated H100 (temperature=0, topk=40, maxtokens=4000), answers from message.content first. Qwen does not report GSM8K on the Qwen3.8-27B card. This score does not apply to the GGUF files in this repo. I’m a PhD student in visual neuroscience at the University of Toronto. Training and release compute is self-funded (rented H100s and a local DGX Spark). If these artifacts are useful, Ko-fi…
Модальности:
Генерация текста Мультимодальность
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: GestaltLabs
Теги: gguf, llama.cpp, quantized, qwen3_5, qwen3.8, multimodal, endpoints_compatible, conversational
Лайков: 4 | Загрузок: 392
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
