Aurora Mobile's GPTBots.ai Integrates Jev — Two Layers of AI, One Enterprise Platform
Aurora Mobile's enterprise AI agent platform GPTBots.ai has integrated Jev, a purpose-built decision model from TypeSafe AI, creating what the company calls a two-layer architecture that separates fast judgments from complex reasoning. Under the setup, Jev handles model routing, retrieval filtering, and intent classification in under 500ms at a fraction of standard LLM cost, while general-purpose models like GPT and Claude handle text generation and open-ended dialogue. GPTBots.ai founder and CEO Chris Lo said the integration changes the economics of the entire pipeline, replacing full LLM calls on every routing decision with sub-cent judgments that carry calibrated confidence scores.
Key figures
- Jev launch date
- 2026-09-15
- Decision layer sla
- sub-500ms routing, filtering, and classification judgments
- Jev decision latency
- 70–500ms end-to-end (vs 3–329 seconds for frontier LLMs, per TypeSafe benchmarks)
- Jev input token cost
- $0.042 per million input tokens; output free (up to 445x cheaper than comparable LLM decision tasks)
- Jev structured output error rate
- 0% vs up to 45.5% for some frontier models (per TypeSafe benchmarks)
AI analysis
The rest of the AI analysis, red flags and sentiment are part of Signal8 Pro.
AI-generated analysis of a public disclosure. Not investment advice; verify against the original document.