Alibaba Releases 2.4T-Parameter Open-Weight Model Qwen3.8-Max
NVIDIAAI · x · 2026-08-13
Alibaba released Qwen3.8-2.4T-A95B (Qwen3.8-Max), its largest open-weight model to date, featuring 2.4T total parameters with 95B activated per token.
- Architecture: A fine-grained Mixture of Experts (MoE) architecture combining full and linear attention, supporting up to a 1M token context window and 128K output length, optimized for reasoning and agentic workloads.
- Inference Performance: According to an NVIDIA technical blog, the model achieves over 4K tokens/second per GPU and 350+ tokens/second per user on the NVIDIA GB300 NVL72 in FP8 precision out of the box.
- Infrastructure: Deploying a 2.4T model requires data-center-scale compute. NVIDIA is collaborating with the open-source ecosystem to provide optimized multi-node deployment recipes.
More from Infra
- MLCommons launches MLPerf Endpoints, a benchmark that measures inference endpoints the way buyers actually purchase them — TheKanter · 2026-10-16
- AI-Driven Timing-Aware Power Optimization Uses Learned Ranking With Baseline-Preserving Repair — Amit Sharma · 2026-10-16
- Nscale AI datacenter plagued by coolant leaks, transformer fire, 10 safety violations — saranormous · 2026-10-16
- WSL 3.0 ships with native containers and Linux 6.18 kernel, syscall overhead down ~16% — solyarisoftware · 2026-10-16
- Are 22 GB modded RTX 2080 Ti cards a viable budget local LLM option? — Dave86ch · 2026-10-16
- Intel's neural-compressor: SOTA low-bit LLM quantization with FP4/NVFP4 support — HaihaoShen · 2026-10-16