A developer demonstrated training Qwen3.8-Flash-Next (125B-A6B plus 51B engram) in 40 GiB VRAM without CPU offload using LoRA over GGUF, reaching 200 token/s on Strix Halo. The method relies on Huggin…
A developer localized the documentation for Strata, an MIT-licensed C++ inference engine that runs a 125-billion-parameter Qwen3.8-Flash-Next model on a single 12 GB consumer GPU via Q2_0 and IQ2_XS l…
VIDRAFT released POCKET-Darwin-180B-GGUF, a 4-bit quantized GGUF build of its 180B-parameter Mixture-of-Experts Darwin-180B-RSI model that runs on consumer hardware, including a gaming laptop with 8 G…
A developer released Juud_engine, a free MIT-licensed experimental fork of Strata v0.1.39 for local Qwen3.8-Flash-Next inference, reporting median paired decode throughput 3.1–9.1% higher and paired f…
Strata v0.1.38 runs Qwen3.8-Flash-Next, a 125B mixture-of-experts model, on a gaming GPU at 94 tokens per second without a server, according to byteiota. The report attributes the feat to the model's …
Ethereum co-founder Vitalik Buterin described a three-layer privacy setup for AI use that runs Alibaba's Qwen3.8-Flash-Next as a local model to rewrite requests before they reach remote systems, combi…
The open-source project Strata, from developer Niko1221, runs a 125-billion-parameter Qwen3.8 mixture-of-experts LLM on gaming-PC hardware by treating VRAM as a cache and keeping the full expert set i…
Open source inference engine Strata, built by a developer known as Niko, runs Alibaba's 125 billion parameter Qwen3.8-Flash-Next mixture of experts model on a single consumer GPU with as little as 8GB…
A homelab test on an i5-12600K with 64 GB DDR5 and an RTX 4070 12GB found the narrow inference runtime Strata generated 512 tokens at 53.2 tok/s with 60,000 tokens of context on the 125B-parameter Qwe…
A community tester published deployment notes for the Halogen Flash Server, an optimized server for running Qwen3.8-Flash-Next on AMD Strix Halo (gfx1151) hardware, reporting a roughly 10% performance…
Korean Pre-AGI startup VIDRAFT released Darwin-180B-RSI, a 180-billion-parameter mixture-of-experts reasoning model built on Qwen3.8-Flash-Next that combines selective model merging with recursive sel…
TensorFold 0.3.6.2 delivered decode speeds over 62 tokens per second on a single stream and 119 tokens per second across five concurrent streams running Qwen3.8-Flash-Next on a single Nvidia DGX Spark…
Strata, an open-source tool from developer Niko1221, runs the 125-billion-parameter Qwen3.8-Flash-Next model on a single NVIDIA GPU with 12-24 GB of VRAM and 64 GB of RAM, generating answers at 60-95 …
A developer got Qwen3.8-Flash-Next running on a 128 GB Mac Studio M4 Max by memory-mapping the model's 32 GB n-gram embedding table from disk on demand, avoiding both the quality loss of the expert-pr…
A September 2026 comparison of open-source coding models found GLM-5.3-Flash from Z.ai best for agentic coding, DeepSeek V4 Flash cheapest per token, and MiniCPM5-2B best for on-device use. GLM-5.3-Fl…
Alibaba released Qwen3.8-Flash-Next on August 26, 2026, an open-weight ultra-sparse mixture-of-experts model with 125B total parameters and roughly 6B active per token, plus a 51B-parameter N-gram emb…
Alibaba's Qwen team released Qwen3.8-Omni-Flash, which it calls its first omni-modal model built around agentic capabilities, accepting text, images, audio, and video and returning text. The model is …
A user running Qwen3.8-Flash-Next at IQ3 quantization on a system with 64GB of RAM, an AMD 9950X3D, a 9070XT, and a ZFS stripe of three mid-range NVMe drives reported fitting 255K tokens of context us…
A new web tool lets users estimate the energy and CO2 emissions of their AI usage, using modeled coefficients from public AgentX benchmarks on NVIDIA B300 hardware. The calculator, posted on Hacker Ne…
Qwen has released Qwen3.8-Flash-Next, an experimental open-weight causal language model with a vision encoder, designed for coding agents, long-horizon tool use, and multimodal computer tasks. The mod…