サイト内検索

AI モデル 84759

• Falcon-H1-Tiny-90M-Instruct↗

  tiiuae

• resnet50d.a3_in1k↗

  timm

• vit_large_patch32_224.orig_in21k↗

  timm

• vit_small_r26_s32_384.augreg_in21k_ft_in1k↗

  timm

• qwen3.6-moe↗

  tiny-random

• Qwen3.5-0.8B-M↗

  trymirai

• gpt2-chinese-poem↗

  uer

• Qwen3.8-Flash-Next-GGUF↗

  vcruz305

  Real K-quant ladder from full HF BF16 weights with **MTP heads in-main** (`blk.48` / `nextn.*`) and a **shared BF16 PLE**. Weights include MTP (`nextn_predict_layers` / `blk.48`). Runtime `--spec-type draft-mtp` needs a llama.cpp build with **qwen4exp graph_mtp** (not all mainline builds yet).

• PhoGPT-4B-Chat↗

  vinai

• d3-turbo↗

  vllm-sr