#DeepSeek

DeepSeek-V5 model identifier appears in Harness code repository, paving the way for the next generation of large models

Developers found three model identifiers—deepseek-v5, deepseek-v5-open, and deepseek-v5-next—in the DeepSeek Harness GitHub repository, currently serving as framework placeholders.

How Inception Mercury 3 beats GPT-6 Luna and Claude Haiku 5.5 with less wait time.

LFM2.5-24B-A2B release: A fast intelligent agent model for device-oriented applications

LFM2.5-24B-A2B is a 24 billion parameter reasoning model that uses only 2 billion active parameters, suitable for fast intelligent agents on devices. The model competes with Qwen3.6-35B-A3B in five key benchmark tests, 1.7 times faster than similar models, and is open source.