{"slug": "model-xing4-0-29b-a4b", "title": "Model: Xing4.0 29B-A4B", "summary": "XingChen-AGI released Xing4.0 29B-A4B, a 29B-parameter mixture-of-experts text model under Apache 2.0 that activates about 4B parameters per token, with GGUF quantizations published by bartowski. The model uses DeepSeek-style MLA attention with a compressed cache of about 46 KB per token, so at Q4_K_M a 32K context window adds only 1.4 GB; Q4_K_M needs about 21.2 GB at 4K, IQ4_XS about 19.0 GB, Q3_K_M about 17.0 GB and Q2_K about 13.6 GB, while the full 262,144-token window at Q4_K_M needs about 33.3 GB. Each GGUF file carries a 1.7B-parameter multi-token-prediction block that llama.cpp skips unless speculative decoding requests it, putting real use about 1 GB under those figures at Q4_K_M, and no per-level quality loss has been published.", "body_md": "A 29B mixture-of-experts text model from XingChen-AGI under Apache 2.0 that reads about 4B parameters per token. Attention is DeepSeek-style MLA with a compressed cache of about 46 KB per token, so context is cheap next to the weights: at Q4_K_M a 32K window adds only 1.4 GB. At Q4_K_M it needs about 21.2 GB at 4K, tight on a 24 GB card. IQ4_XS needs about 19.0 GB at 4K, comfortable on a 24 GB card. Q3_K_M needs about 17.0 GB at 4K, comfortable on a 20 GB card. Q2_K needs about 13.6 GB at 4K, comfortable on a 16 GB card. The full 262,144-token window at Q4_K_M needs about 33.3 GB at 256K. Each GGUF file also carries a 1.7B-parameter multi-token-prediction block that llama.cpp skips unless speculative decoding asks for it, so real use is about 1 GB under these figures at Q4_K_M. Sizes come from bartowski's GGUF files, and no per-level quality loss has been published.", "url": "https://wpnews.pro/news/model-xing4-0-29b-a4b", "canonical_source": "https://quantized.uk/quant-hub/xing4-0-29b-a4b/", "published_at": "2026-11-10 12:00:00+00:00", "updated_at": "2026-11-11 01:20:24.266503+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-research", "ai-tools"], "entities": ["XingChen-AGI", "Xing4.0 29B-A4B", "bartowski", "llama.cpp", "DeepSeek"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/model-xing4-0-29b-a4b", "markdown": "https://wpnews.pro/news/model-xing4-0-29b-a4b.md", "text": "https://wpnews.pro/news/model-xing4-0-29b-a4b.txt", "jsonld": "https://wpnews.pro/news/model-xing4-0-29b-a4b.jsonld"}}