Arcee AI launches Trinity: a 400-billion-parameter open source LLM Arcee AI, a 30-person startup, has developed Trinity, a 400-billion-parameter open source LLM, challenging the tech giants.
LLMs and the Threat to Online Privacy: The End of Pseudonymity Large language models (LLMs) are threatening online privacy, making pseudonymity obsolete as a protective measure.
Skillware: Deterministic Middleware for LLM Prompt Compression Learn how Skillware, a deterministic middleware, can compress LLM prompts to cut costs and improve efficiency.
Optimizing LLM Inference with NVIDIA KVPress Learn how NVIDIA KVPress optimizes inference for long-context language models, improving efficiency and reducing memory usage.
Gemma 4: open models for reasoning and agents Gemma 4 brings reasoning, agentic workflows, multimodality and flexible deployment across edge, workstations and cloud for enterprise use cases.
Sycophancy in LLMs: How It Is Measured and Mitigated In-depth analysis of sycophancy in large language models, measurement methods and mitigation strategies for more reliable and ethical models.