A developer published a serving configuration for the turboderp/Qwen3.8-Flash-Next-exl3-4.05bpw EXL3 quant, running the large MoE model split across an RTX 5090 and RTX 4090 with 192 GB of DDR5 system…
A developer roundup highlights four open-weight coding models for local workstations in late 2026, including Meta Muse Spark 1.2, Alibaba Qwen3.8-27B, Meta Muse Glimmer 30B, and Alibaba Qwen3.8-Flash-…
Qwen3.8-Flash-Next, a sparse Mixture-of-Experts model with roughly 125B total parameters but only about 6B active per token, offers a preview of Qwen's architectural direction for Qwen4. The model emp…
A developer's analysis of recent Hacker News threads suggests the local AI ecosystem is becoming a practical reality, with users buying Macs specifically to run models locally and successfully running…
Alibaba's Qwen team released Qwen3.8-Flash-Next on August 26, 2026, an open-weight experimental model with 125B total parameters but only 6B active per token, previewing the architecture intended for …
Independent developer Carlos Galarza released slotstream, a Swift binary that streams Mixture-of-Experts model tensors from NVMe SSD into Apple Silicon Metal memory, enabling local inference of Alibab…
Slotstream, a new Swift binary, lets users run the 104 GB Qwen3.8-Flash-Next model on a 48 GB Mac at about 12 tokens per second by streaming weights from SSD, with peak memory of 32 GB and a cold star…
Z.ai released GLM-5.3-Flash, a 320B-parameter multimodal MoE model with 18B active parameters, and Alibaba's Qwen team released Qwen3.8-Flash-Next, a 125B model with 6B active parameters, within a day…
A developer has published a configuration guide for running the Qwen3.8-Flash-Next 125B MoE model with llama.cpp on an RTX 4090 24GB system, achieving up to 29 tokens per second decode speed. The setu…
A developer documented a runbook for deploying the Qwen 3.8 Flash Next model on a single NVIDIA Blackwell GPU using a specialized vLLM runtime. The deployment supports both Docker and native systemd t…
Z.ai revealed that the mystery model Ox-Alpha on OpenRouter and OpenCode was its GLM-5.3-Flash, a 320B-parameter multimodal MoE model with 18B active parameters, and released its weights on Hugging Fa…
Qwen released Qwen3.8-Flash-Next, a 125B-parameter open-weight multimodal MoE model built on the Qwen4 architecture with a 262K context window, which outperforms Claude-4.6-Opus (Max) and can run loca…
Alibaba's Qwen3.8-Flash-Next, released on August 26, 2026, scores 56 on the Artificial Analysis Intelligence Index, well above the median of 28 for comparable open-weight models, and is priced at $0.1…
Alibaba released a preview of Qwen3.8-Flash-Next, a 125-billion-parameter model that activates only 6 billion parameters per token, achieving training costs roughly one-ninth of typical models of its …
Four papers published today challenge the reliability of LLM-as-judge, showing that judge annotations can mark working code as failures, leak across rubric dimensions, and misalign with human creativi…
Apple Inc. released new Mac mini and Mac Studio models targeting AI inferencing, with the Mac Studio featuring an M5 Ultra chip and 256GB of RAM priced at AUD $17,500, and the Mac mini with an M5 Pro …
A developer tested the Qwen3.8-Flash-Next model on an NVIDIA DGX Spark (GB10, 128 GB unified memory) using llama.cpp with quantized KV cache, achieving 128K context and 414 tok/s prefill. They identif…
Alibaba released the model weights for Qwen3.8-Flash-Next, a multimodal mixture-of-experts model with 125B parameters plus 51B N-gram embeddings and 6B activated per token, as a preview of the Qwen4 a…
Alibaba released the model weights for Qwen3.8-Flash-Next, a multimodal mixture-of-experts model with 176B total parameters, as a preview of the upcoming Qwen4 architecture. NVIDIA provides Day 0 func…
Alibaba's Qwen organization activated the Qwen3.8-Flash API on QwenCloud on August 26, 2026, pricing it at $0.16 per million input tokens and $0.47 per million output tokens with a default 1-million-t…