Qwen3.8-KAT-AgentWorld-35B-A3B-Coder
INVALID LANGUAGE PAIR SPECIFIED. EXAMPLE: LANGPAIR=EN|IT USING 2 LETTER ISO OR RFC3066 LIKE ZH-CN. ALMOST ALL LANGUAGES SUPPORTED...
INVALID LANGUAGE PAIR SPECIFIED. EXAMPLE: LANGPAIR=EN|IT USING 2 LETTER ISO OR RFC3066 LIKE ZH-CN. ALMOST ALL LANGUAGES SUPPORTED...
1. Layer replacement. The 8 full-attention layers listed above were replaced with GDN/linear-attention layers, each initialized from a...
OdinNext is a 138M-parameter causal language model that replaces softmax self-attention with an HGRN2-style gated linear recurrence. This...
A 2-billion-parameter dense hybrid language model pretrained from scratch on 500B–2T tokens. BOREAL-2B is the first model in...
GDN-primed-HQwen3-32B-Instruct is a Hybrid language model consisting of 50% Attention layers and 50% Gated DeltaNet (GDN) layers, primed...
GKA-primed-HQwen3-8B-Reasoner — это гибридная языковая модель, состоящая из 50% слоев внимания и 50% слоев стробированного KalmaNet (GKA), загрунтованных...
RWKV-GLM-4.7-Flash-exp is an alpha-stage experimental model that converts GLM-4.7-Flash into a fully linear-attention-dominant architecture using the RADLADS distillation...
Qwen3.5-REAP-20B-A3B is a Mixture of Experts (MoE) model created by applying Router-weighted Expert Activation Pruning (REAP) to Qwen/Qwen3.5-35B-A3B....
Кванты включения AI/Ling-3.0-flash, 127,5В-общая/5,1В-активная гибридная МО (bailinghybrid/BailingMoeV3`). Каждый файл здесь содержит только типы тензоров, безопасные для sm120 —...
Официальная сборка OsaurusAI MXFP8 Qwen/Qwen-AgentWorld-35B-A3B (Apache-2.0) — гибридный MoE ~35B / ~3B (qwen35moe). 8-битный микромасштабируемый FP практически без...