Microsoft AI Releases MAI-Cyber-1-Flash: A 5B-Active-Parameter Cyber Model That Pushes MDASH to 95.95% on CyberGym
Microsoft AI has released MAI-Cyber-1-Flash, its first model built specifically for cyber defense. The model does not ship as a standalone endpoint. It runs inside MDASH, Microsoft’s multi-model agentic scanning harness. MAI-Cyber-1-Flash MAI-Cyber-1-Flash is a transformer with self-attention and sparse Mixture-of-Experts layers. It carries 137B total parameters with 5B active, and a 256k context length. Inputs and outputs are text only. It is a cybersecurity-specialized fine-tu
Get smart on it
Microsoft AI has released MAI-Cyber-1-Flash, a specialized AI model designed to find software vulnerabilities as part of a larger system called MDASH. The model contains 137 billion total parameters with 5 billion active parameters and is built on technology already used in coding tools. When MDASH uses MAI-Cyber-1-Flash alongside another model on a benchmark of real-world vulnerability tasks, it achieves a 95.95% success rate, representing a significant improvement over previous configurations. The model is designed defensively to help patch bugs rather than create exploits, and it handles most routine security scanning tasks while sending only the hardest 10% of work to a larger model, reducing overall costs.
Related news
Pokee AI Releases Pokee-Isaac 28B: A 10M-Token Context Agentic Model Built to Run Inside the Customer Boundary
Long-horizon agents accumulate context faster than they resolve tasks. Every tool output, observation, and intermediate reasoning step stays in the window, and the two capabilities that matter — holding that context and staying coherent across it — have so far been available almost exclusively from cloud endpoints. That excludes regulated industries, public-sector institutions, and on-device applications, where the data is not permitted to leave the boundary at all. Pokee AI released Pokee-Isaa
Mistral AI Releases Shieldstral 1.0 3B: An Open-Weights Policy-Adaptive Multimodal Safety Classifier Matching Models 7× Its Size
Mistral AI has released Shieldstral 1.0 3B, an open-weights, policy-adaptive multimodal safety classifier that treats content moderation as a single yes/no question rather than a fixed taxonomy of harm categories. Most guardrail models bake their category list into the weights, so re-targeting one to a new deployment context means retraining — and the same content can be acceptable on a cybersecurity research tool while being harmful on a mental-health platform. Shieldstral inverts that: operat
