Qwen3.8-27B-Uncensored-exl3-4bpw
EXL3 квантование orcarouter/Qwen3.8-27B-Uncensored, аблитерированная (отказ-удаленная) сборка Qwen/Qwen3.8-27B. 16 ГБ на диске — подходит для одной карты 24 ГБ...
Метка: qwen3_5
O2-9b-preview
> 9B агентская модель, обученная на основе взаимодействий, сгенерированных в средах выполнения, развернутых в реальных рабочих процессах офиса...
Ornstein3.8-27B-GGUF
GGUF quantizations of GestaltLabs/Ornstein3.8-27B: a Qwen 3.8 27B dense vision-language fine-tune with interleaved linear and full attention (Gated...
Qwen3.8-27B-Della-Deckard-Fable-Qwopus-ColdFusion-v4
三阶段合成。第一阶段用 DELLA 把三个不同渊源的模型融成一个“通用智能复合体 G”, 第二阶段把 Qwopus3.6-27B-Fusion 相对祖先的增量原封不动地线性叠加到 G 上。 两阶段的 basemodel 均为真祖先 Qwen/Qwen3.6-27B(不是任何一个子代),因此所有 任务向量 Δ 都在同一个参考系里,叠加才有意义。 第三阶段把前两阶段的全部成果整体换锚:out =...
grug-v1.1-qwen-3.8-27b-mtp
same grug as ProCreations/grug-v1.1-qwen-3.8-27b, plus a draft head that guess ahead, tuned on grug own output. Qwen3.8 ship...
Qwen3.8-27B-abliterated
Abliterated (uncensored) version of Qwen/Qwen3.8-27B with the refusal direction removed from the model weights, allowing unrestricted conversations. Abliteration...
AX-Qwen3.8-27B-MLX-AXQ-4bit-MTP
An AXQuant (AXQ) mixed-precision MLX checkpoint for Apple Silicon, converted directly from the BF16 source model. The language...
Qwen3.6-40B-Fable-Fusion-6-Core-Deckard-Eleanor-Heretic-Uncensored-W4A16-AutoRound
4-bit weight-only quantization (W4A16) of DavidAU’s 40B «Chimera» — the 6-Core Fable Fusion Heretic Uncensored model. Quantized with...
Qwen3.5-2B
LiteRT is Google’s on-device runtime, the new name for TensorFlow Lite (Android: com.google.ai.edge.litert:litert), and litert-torch, the renamed ai-edge-torch,...
HybridDiffusion-9B
HybridDiffusion-9B is an 8.95B-parameter text-generation checkpoint produced by applying HybridDiffusion’s autoregressive-to-diffusion conversion recipe to Qwen3.5-9B. A single set...