open-spaces

Overview of open-spaces

What it is

Jev is TypeSafe AI's first System One model, designed for software that needs a judgment rather than a paragraph. It accepts text or JSON state plus typed questions and returns constrained answers with probabilities, which ordinary code can branch on. Jev is used to classify, route, score, detect, rank, extract, verify, and gate automation, keeping final control flow, thresholds, and side effects in code. This is an independent community collection, not an official TypeSafe AI repository. The snapshot was reviewed on September 19, 2026.

Key features

• Input: a string, JSON object, or array of text values - Output: Choice, Score, and Noul answers with typed fields; Choice and Score also include probabilities and confidence - Current listed price: $0.042 per 1M input tokens; output tokens are free - Current listed limits: 250,000 tokens/second and 1,200 requests/minute; context length 64k tokens per request - End-to-end response times reported by TypeSafe: 70–500 ms (vendor-reported, workload-dependent)
• Primary language: English; test non-English workloads on representative data before relying on calibrated thresholds - Data handling: customer requests and responses are not used to train models - Modalities: text only; images, audio, and video are not supported

Use cases

Jev fits the fuzzy middle between brittle rules and expensive generative workflows. Use it to:

• Pick one value from a known set (Choice)
• Measure a position on an ordered rubric (Score)
• Estimate whether a condition is true (Noul)
• Decide whether to act, review, or fall back (probabilities + confidence + code)
• Write an explanation, code, or a reply (use an LLM, optionally after Jev routes or verifies it)

Getting started

See the repository README for installation and quick-start commands.

License and pricing

The project is licensed under MIT. The Jev API is priced at $0.042 per 1M input tokens; output tokens are free.

Based on the project's public README and repository metadata.