Records found: 5
'Alibaba Qwen released compact dense Qwen3-VL 4B and 8B models in Instruct and Thinking variants with FP8 checkpoints for low-VRAM deployment, retaining the same multimodal capabilities and long-context support.'
'Meta FAIR released CWM, a 32B dense decoder-only LLM mid-trained on Python execution traces and agent interactions to teach semantics and long-horizon reasoning for code generation.'
Alibaba's Qwen team published FP8-quantized checkpoints for Qwen3-Next-80B-A3B (Instruct and Thinking), enabling more efficient long-context inference and practical deployment of an 80B/3B-active hybrid MoE on commodity GPUs
'IBM published two ModernBERT-based Granite R2 embeddings offering 8k context, compact architectures, and high retrieval throughput suitable for production RAG and search systems.'
'Baidu released ERNIE-4.5-21B-A3B-Thinking, a 21B-parameter MoE model with 3B active parameters and 128K context length designed for efficient multi-step reasoning.'