Hy3-295B-A21B-prismaquant-gridbook-2.9bit-vllm
A 295B-A21B MoE served on one 128 GB DGX Spark at 2.9 bits per parameter, through stock vLLM...
A 295B-A21B MoE served on one 128 GB DGX Spark at 2.9 bits per parameter, through stock vLLM...
This is a single-download, two-node GB10 serving bundle. It combines an offline-prepared NVIDIA NVFP4 W4A4 target with the...
Inferact/MiniMax-M3-EAGLE3-GQA — это черновая модель EAGLE3 с групповым запросом и вниманием (GQA) для ускорения вывода MiniMax-M3, обслуживаемая с...
This is a second pass at quantizing Gemma 4 31B IT to NVFP4. My earlier upload (gemma-4-31B-it-qat-NVFP4-Blackwell) used...
This repository contains a 4-bit RTN-quantized version of the Tencent Hy3 Mixture-of-Experts model. Hy3 is a 295B-parameter MoE...
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...
Calibrated NVFP4 quantization of deepreinforce-ai/Ornith-1.0-35B (Qwen3.5-35B-A3B hybrid MoE) for vLLM. 24 GB, and it decodes at the same...
> [!WARNING] > This model is a beta preview and a research artifact. It is not an open...
kimi-k2.7-code-eagle3-mla is an Eagle3 MTP draft model with MLA (Multi-Latent Attention) for accelerating inference of Kimi-K2.7-Code under vLLM...
Квантование NVFP4 (W4A4) модели мышления (рассуждения) llm-jp/llm-jp-4-32b-a3b-thinking — японской модели мышления (рассуждения) Mixture-of-Experts с 32B-A3B битами. Квантовано с...