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-Isaac 28B, a 28B text-only foundation model with a 10M-token context window, designed to run inside that boundary. The Pokee research team claims 93.3% on RULER at 10M tokens, parity with the strongest cost-optimized cloud baselines on agentic benchmarks, and a serving profile that fits a single GPU.
Is it deployable
Yes — but licensed, not open-weight. Pokee AI serves Isaac through an OpenAI-compatible developer API, and licenses it for deployment inside a VPC, on-premises, or on-device. The launch announcement advertises Day-0 support for vLLM and SGLang, and single-GPU serving starting from an RTX 4090 or equivalent. The research team publishes measurements only from a single B200-class GPU, so treat the consumer-GPU claim as vendor guidance rather than a reported result.
Company level: This fits organizations that already own their inference stack — mid-size and enterprise teams with a platform group, plus device OEMs. A solo practitioner without on-prem hardware should use the hosted API instead; the boundary argument only pays off if you have a boundary.
Industries: Healthcare and payors, financial services and insurance, defense and public sector, legal and e-discovery, and pharma or semiconductor R&D. The common trait is a rule that the data cannot cross an external API boundary, not a preference for privacy.
Applications: Whole-repository code review, multi-year contract and claims analysis, incident forensics over full log archives, and long-running tool agents that never need summarization or context pruning. The research paper makes this second point explicitly: when enough usable context is available in-boundary, memory hierarchies and compression become optional rather than required.