LFM2.5-8B-A1B-SOMPOA-heresy
This is an LFM2.5-8B-A1B fine-tune, produced through P-E-W’s Heretic (v1.4.0) abliteration engine with Self-Organizing Maps & Magnitude-Preserving Orthogonal...
This is an LFM2.5-8B-A1B fine-tune, produced through P-E-W’s Heretic (v1.4.0) abliteration engine with Self-Organizing Maps & Magnitude-Preserving Orthogonal...
A decensored (Abliterix-abliterated) version of deepreinforce-ai/Ornith-1.0-35B — a Qwen3.5-MoE 35B coding and reasoning model (256 experts/layer, 8 active...
> [!TIP] > Эта модель воспроизводима! > > Смотрите README в каталоге воспроизведения для получения дополнительной информации. userinput2...
> [!TIP] > This model is reproducible! > > See the README in the reproduce directory for more...
The method invented by «P-E-W» looks for the best settings to de-censor («abliterate») the model by trial and...
INVALID LANGUAGE PAIR SPECIFIED. EXAMPLE: LANGPAIR=EN|IT USING 2 LETTER ISO OR RFC3066 LIKE ZH-CN. ALMOST ALL LANGUAGES SUPPORTED...
LFM2.5-350M is a general-purpose text-only model with the following features: — Number of parameters: 350M — Number of...
NVFP4 (W4A4) quantized version of trohrbaugh/GLM-4.7-heretic — a decensored zai-org/GLM-4.7, made using Heretic v1.2.0+custom. > Requires NVIDIA Blackwell...
—————————-|:——-:|:——-:|:—————-:|:————-:|:—————:|:——————:|:———-:|:————:| > Coding: AGI is a long journey, and benchmarks are only one way to evaluate performance. While...
Сегодня мы анонсируем Qwen3-Coder-Next, языковую модель с открытым весом, разработанную специально для агентов кодирования и локальной разработки. Он...