Qwen3.8-Flash-Next Multimodal MoE Model: ~402GB Memory
Explore Qwen3.8-Flash-Next, a next-gen open-weight multimodal causal language model featuring hybrid attention, MoE a ...
LFM2.5-350M-Diffusion-Exp Text Generation Model: 4GB+ VRAM
LiquidAI has released LFM2.5-350M-Diffusion-Exp, an experimental block-diffusion model converted from its lightweight ...
GLM-5.3-Flash-GGUF Vision-Language Model: ~150GB Memory
Discover ggml-org/GLM-5.3-Flash-GGUF, a GGUF quantization of zai-org’s native multimodal MoE model, featuring v ...
Aleph Alpha Releases Kolibri: A New Open-Weight MoE Model
German AI firm Aleph Alpha released Kolibri, an open-weight Mixture-of-Experts model supporting English and German wi ...
Magnitude CLI v0.2.5 Released: M5 Mac & MoE Speedups
Version 0.2.5 of @magnitudedev/cli is out, bringing 2x faster prompt processing on M5 Macs, MoE speedups, and M1/M2 f ...
LocalAI v4.11.0 Released with Failover and NeMo Audio Models
LocalAI v4.11.0 introduces model failover chains, NeMo speech-to-text models, decision models, signed OCI galleries, ...
Magnitude 0.2.4 Released: Faster Inference and Lower Memory
Magnitude 0.2.4 is released with significantly reduced memory usage, faster kernel tuning, and improved error handlin ...
Ai2 Releases AstaBrief 8B: Fast Open-Source Scientific Report…
Allen Institute for AI releases AstaBrief 8B, a fast open-source scientific report generation model with training dat ...
ELYZA Releases ELYZA-Thinking-1.0 32B/33B Reasoning Models
ELYZA has released ELYZA-Thinking-1.0, a reasoning model based on the llm-jp-4 series, featuring strong Japanese-Engl ...
SGLang v0.5.21 Released with Dynamic PD Role Switching
SGLang v0.5.21 is out, bringing dynamic PD role switching, a Rust-based prefix cache core, new APIs, and expanded mod ...