Jev and openjev are not the same thing
openjev is a community effort to reproduce the idea on hardware you own. It is not a client for Jev, and it is not built by TypeSafe.
01What openjev actually is
openjev is not a client for Jev, not an open release of Jev, and not a TypeSafe project. It is an independent effort that started within days of the Jev launch, asking whether the same idea — typed probabilistic answers instead of generated text — can run on hardware you already own.
The main repository is TheoLeeCJ/openjev, written in Python, and its stated question is whether something like Jev can run on a single consumer GPU at home. A second project, ekzhang/openjev-sglang, exposes a Jev-compatible endpoint backed by open models, which is the closest thing to a drop-in local substitute.
02How they differ
| Jev | openjev | |
|---|---|---|
| Built by | TypeSafe AI | Independent contributors |
| Runs | Hosted API only | On your own machine |
| Weights | Not published | Open models underneath |
| Costs | $0.042 per million input tokens | Your hardware and power |
| Maturity | Production service | Days old, moving fast |
03When a local reproduction makes sense
If the data cannot leave your network, if you need to classify volumes where per-token pricing stops being trivial, or if you want to study how the approach works, a local reproduction is worth the setup. The tradeoff is that you own the quality problem: nothing guarantees a local cross-encoder is calibrated the way a hosted model claims to be, and calibration is the whole value of a probability.
04When it does not
For most applications the hosted model is cheaper than the GPU you would buy to avoid it. At the published rate, a million judgments costs about fifteen dollars, which is less than a day of the instance you would rent to serve them yourself.