Skip to content

Alibaba Cloud· Released Aug 26

Research made with Qwen3.8-Flash-Next

Benchmarks, write-ups and discussions.

GVS5H: five Qwen3.8 models matching a frontier model on LiveCodeBench HardAug 26

Entries

David Hendrickson

@TeksEdge · Sep 26

This guy was getting ~15 tps running Qwen3.8-Flash-Next on his 12GB RTX 5070.

Apparently that wasn't good enough. 😂

So he built his own inference engine. (as we all should)

Now he's reporting
🐌 llama.cpp → ~15 tps
🚀 Strata → up to 65.1 tps

And this isn't big workstation…

0xSero

@0xSero · Sep 29

GPU poor rejoice!

You will be able to run Qwen3.8-Next-Flash on a single 24-32 GB GPU + 64GB ddr4/ddr5 + 100GB NVMe/SSD

RTX 3090:

  • 2000 / 2800 tok/s prefill
  • 64 tok/s decode c1
  • 210k fp8 cache

Intel Arc B70:

  • 800 / 1200 prefill tok/s
  • 35 decode tok/s
  • 270k fp8 cache

Unsloth AI

@UnslothAI · Aug 26

Qwen3.8-Flash can now be run locally! 🔥

The 125B MoE model outperforms Claude-Opus-4.6 (Max).

Run on 75GB RAM via Unsloth GGUFs.

Qwen3.8-Flash-Next enables CPU RAM / unified mem setups to deliver near VRAM speeds.

Guide: https://t.co/LLMclyJTeL
GGUF:

ハカセ アイ

@ai_hakase_ · Sep 26

【Qwen3.8-Flash-Nextを12GB VRAMで激速化する「Strata」】
👉 https://t.co/H6Cz6pixg0

12GB VRAMの限定された環境でも、カスタム推論エンジン「Strata」と高度な非一様量子化を使うことで、Qwen3.8-Flash-Nextを最高クラスの速度で動かせるそうです！なんと出力が約65 tok/sという驚異的なスピードですね！✨

遅延…