Pokee AI Releases Pokee-Isaac 28B: A 10M-Token Context Agentic Model Built to Run Inside the Customer Boundary
Pokee AI launched Pokee-Isaac 28B, a 28-billion-parameter foundation model featuring a 10-million-token context window designed for private deployment. The model reportedly achieves 93.3% on RULER benchmarks at that length.
Pokee AI has introduced Pokee-Isaac 28B, a new text-only foundation model designed specifically for enterprise environments requiring strict data governance. The release highlights a focus on agentic capabilities within a controlled deployment infrastructure.
The model boasts a 10-million-token context window, significantly exceeding standard limits. According to the company, it scores 93.3% on the RULER benchmark at this length, while comparison baselines reportedly fail beyond 2 million tokens.
Performance metrics also include a 70.94 score on BFCL v4, indicating strong function calling capabilities. These specifications target complex workflows that require processing large volumes of data without external data leakage.
By prioritizing deployment inside the customer boundary, Pokee AI addresses security concerns prevalent in regulated industries. This approach positions the model for use cases where data sovereignty and long-context reasoning are critical requirements.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.