TGTGInsighttelegram intelligenceLIVE / telegram public index
← Devils Below

TGINSIGHT SIMILAR POSTS

Find similar content

Source channel @devilsbelow · Post #445 · Jan 13

🌐Weekly News Digest [ January 5 – January 11 ] That was the first full-fledged week of the new year of 2026, during which rulers forgave those who polluted their land, dismissed those who were managing their oil. 💡Here are the key highlights: 🇧🇼 Botswana — Botswana Invites Russia to Invest in Its Mining Sector 🇨🇩 DR Congo — Congolese clergy speaks against the US-DRC agreement. — The government allows processing units to accept ore from artisanal miners amid protests 🇬🇶 Equatorial Guinea — Equatorial Guinea moves its capital to a brand new city built on oil revenues 🇬🇭 Ghana — A Ghanaian prophet predicts the discovery of major onshore oil deposits in Ghana — Ghana hopes to keep its oil fields viable until 2040 🇲🇱 Mali — JNIM militants attack a gold mine in southeastern Mali 🇳🇪 Niger — Niger replaces its oil minister 🇳🇬 Nigeria — President reshuffles the country's oil sector management 🇸🇩 Sudan — Sudan’s central bank and Sudanese Mineral Resources Company set up a joint commission to curb illegal gold exports. 🇺🇬 Uganda — Uganda to start its first oil exports by October, despite environmental concerns 🇿🇲 Zambia — First report on the toxic pollution caused by a Chinese company designates 160 people as victims — Zambia is concerned over the safety of its workers in southern DRC #NewsDigest ➡️ Follow to stay informed - @devilsbelow

Hashtags

Results

1 similar post found

Search: #efficientmodels

当前筛选 #efficientmodels清除筛选
Machinelearning

@ai_machinelearning_big_data · Post #8519 · 09/11/2025, 06:21 PM

🚀 Релиз:Qwen3-Next-80B-A3B - эффективная модель заточенная на работа работу с очень длинным контекстом! 🔹80B параметров, но активируется только 3B на токен → тренировка и инференс 10x дешевле и быстрее, чем у Qwen3-32B (особенно при 32K+ контексте). 🔹Гибридная архитектура: Gated DeltaNet + Gated Attention → сочетает скорость и точность. 🔹Ultra-sparse MoE: 512 экспертов, маршрутизируется 10 + 1 общий. 🔹Multi-Token Prediction → ускоренное speculative decoding. 🔹 По производительности обходит Qwen3-32B и приближается к Qwen3-235B в рассуждениях и long-context задачах. 🟢Qwen3-Next-80B-A3B-Instruct показатели почти на уровне 235B flagship. 🟢Qwen3-Next-80B-A3B-Thinking превосходит Gemini-2.5-Flash-Thinking. ▪Попробовать: https://chat.qwen.ai ▪Анонс: https://qwen.ai/blog?id=4074cca80393150c248e508aa62983f9cb7d27cd&from=research.latest-advancements-list ▪ HuggingFace: https://huggingface.co/collections/Qwen/qwen3-next-68c25fd6838e585db8eeea9d ▪ ModelScope: https://modelscope.cn/collections/Qwen3-Next-c314f23bd0264a ▪Kaggle: https://kaggle.com/models/qwen-lm/qwen3-next-80b ▪ Alibaba Cloud API: https://alibabacloud.com/help/en/model-studio/models#c5414da58bjgj @ai_machinelearning_big_data #AI#LLM#Qwen#DeepLearning#MoE#EfficientModels#LongContext#Reasonin