LLM EXPLORER 65,121 MODELS INDEXED

Qwen3.8 Flash Next GSQ RCO Q2 MTPLX by trustos

By trustos · 350 downloads
HF Repo ↗ Copy URL

Qwen3.8 Flash Next GSQ RCO Q2 MTPLX is an open-source language model by trustos. Features: 126.2b LLM, VRAM: 4.9GB, License: other, Quantized, LLM Explorer Score: 0.3.

  Arxiv:2604.18556   Arxiv:2605.00649   2-bit   4-bit   Apple-silicon Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-flash-...   Conversational   Flash-next   Gsq   Gsq-rco   Image-text-to-text   Ist-daslab   Local-llm   M5-max   Macos   Mixed-precision   Mlx   Moe   Mtp   Mtplx   Multi-token-prediction   Multimodal   Q2   Quantization   Quantized   Qwen   Qwen3.8   Qwen3.8-flash-next   Qwen4   Qwen4 exp   Rco   Region:us   Safetensors   Sharded   Speculative-decoding   Tensorflow   Vision
Model Card on HF 🤗: https://huggingface.co/trustos/Qwen3.8-Flash-Next-GSQ-RCO-Q2-MTPLX 

Qwen3.8 Flash Next GSQ RCO Q2 MTPLX Parameters and Internals

LLM NameQwen3.8 Flash Next GSQ RCO Q2 MTPLX
Repository 🤗https://huggingface.co/trustos/Qwen3.8-Flash-Next-GSQ-RCO-Q2-MTPLX 
Base Model(s)  Qwen/Qwen3.8-Flash-Next   Qwen/Qwen3.8-Flash-Next
Model Size126.2b
Required VRAM4.9 GB
Updated2026-10-21
Maintainertrustos
Model Typeqwen4_exp
Model Files  0.6 GB: 1-of-3   4.3 GB: 2-of-3   0.0 GB: 3-of-3   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB   0.8 GB
Quantization Typeq2
Model ArchitectureQwen4ExpForConditionalGeneration
Licenseother
Transformers Version5.8.0.dev0

Best Alternatives to Qwen3.8 Flash Next GSQ RCO Q2 MTPLX

Best Alternatives
Context / RAM
Downloads
Likes
...sh Next Abliterated MTPLX 4bit0K / 80.5 GB17671