Qwen3.8-27B-MTPLX-bf16
> Вес MTP сохранен. Это полноточная сборка MLX — никакого квантования. Головка мультитокен-предсказания сохраняется на уровне bf16 (mtppolicy:...
> Вес MTP сохранен. Это полноточная сборка MLX — никакого квантования. Головка мультитокен-предсказания сохраняется на уровне bf16 (mtppolicy:...
A Q8_0 quantized GGUF of Qwen3.5-0.8B fine-tuned on 685 distilled question/answer pairs harvested from the Chipotle support chatbot...
HybridDiffusion-2B is a 2.39B-parameter text-generation checkpoint produced by applying HybridDiffusion’s autoregressive-to-diffusion conversion recipe to Qwen3.5-2B. A single set...
Four GGUF checkpoints for llama.cpp, from 3.21 GB to 9.53 GB. Start with M: 5.06 GB and ≈100%...
This repository contains an oQe6 mixed-precision MLX quantized version of Qwythos-9B, optimized for efficient local inference on Apple...
Abliterated GGUF quant ladder of OpenMOSE/Qwen3.5-REAP-262B-A17B, itself a 34% REAP expert-pruning of Qwen3.5-397B-A17B down to 262B total /...
Qwen3.5 27B Dflash, self-quantized to GGUF by Atomic Chat. Built straight from Z Lab’s original weights with a...
> [!TIP] > Multi Token Prediction (MTP) weights are restored from the base model (Qwen3.5-9B). > Qwythos-9B is...
We reduce Qwen3.5-4B inference latency on a single A10G by quantizing both the target model and its speculative-decoding...
Ornith-1.0-9B-DSpark is a production-qualified DSpark speculative-decoding draft for deepreinforce-ai/Ornith-1.0-9B. This repository contains the draft/speculator only. It is not...