Tag: #moe

Huawei openPangu 505B open source model cover

Model Introduction August 7, 2026

openPangu-2.0-Pro: Huawei's 505B Ascend-Native LLM

Huawei releases openPangu-2.0-Pro — a 505B parameter open-weight MoE model trained entirely on Ascend 910B NPUs. Architecture breakdown, hardware sovereignty implications, and honest assessment.

#huawei #openpangu #open-source-llm #moe #ascend #ai-agent

DeepSeek V4 Flash 0731 model introduction showing MoE architecture and agent benchmarks

Model Introduction August 5, 2026

DeepSeek V4 Flash 0731: $0.28 Beats Its Own Flagship

DeepSeek V4 Flash 0731 exits preview with 82.7 Terminal-Bench and 54.4 DeepSWE. 284B/13B MoE, 1M context, MIT license, at $0.14/$0.28 per million tokens. Agent benchmarks beat V4-Pro-Preview.

#deepseek #deepseek-v4-flash #open-source-llm #moe #ai-agent #coding-agent

Qwen 3.8 Max model introduction cover showing Alibaba's 2.4T multimodal flagship LLM

Model Introduction August 3, 2026

Qwen 3.8 Max: Alibaba's 2.4T Multimodal Flagship

Qwen 3.8 Max is Alibaba's 2.4 trillion parameter multimodal flagship with 1M context, MoE architecture, and top Arena.AI rankings. Specs, pricing, and how to access it on SandBase.

#qwen-3.8 #alibaba #multimodal #ai-agent #frontier-llm #moe