An open-source alternative to Jev.
Noma is an open-weight decision model you host yourself. It does the same job as Jev, speaks the same /v1/systemone API, and answers in about 16 ms. This page compares the two honestly, including where Jev is the better choice.
FourreasonsteamsmoveoffahosteddecisionAPI.
Jev is a hosted API. Noma is one 5.4 GB file you run on your own GPU, so states never leave your infrastructure and there is no per-call fee.
A hosted call includes a network round trip. Running next to your agent, Noma answers in 16 ms at the median on one H100.
Noma's weights, serving code, trainer and evaluation are public under MPL-2.0. You can inspect, fine-tune and redistribute them.
Every Noma answer carries a separate abstain probability and an ensemble uncertainty estimate, alongside the option probabilities.
Noma,Jevandotherdecisionmodels,sidebyside.
| Model | Weights | How you run it | Easy | Standard | Hard | Median latency | Cost per 1,000 |
|---|---|---|---|---|---|---|---|
| NomaBlackdrome AI Labs | Open, MPL-2.0 | Self-hosted, one 5.4 GB file | 100% | 98.6% | 51.4% | 16 ms | $0.023 |
| Jev 1.13.0TypeSafe | Closed | Hosted API | 100% | 99.0% | 74.1% | 652 ms | $0.040 |
| OpenJev (thinking)Community | Open | Self-hosted, 26B MoE | 100% | 100% | 78.2% | 463 ms | not listed |
| Nimble 9BCommunity | Open | Self-hosted, Qwen3.5-9B | 100% | 94.8% | 65.5% | 389 ms | $0.166 |
| CygnetCommunity | See leaderboard | Frozen Gemma-4-12B | 100% | 96.9% | 75.5% | 35 ms | $0.037 |
| decider-4b v2Community | See leaderboard | 4B model | 100% | 96.9% | 67.3% | 17 ms | $0.020 |
Noma's figures are our own runs with the JevBench client on one H100 and are awaiting independent measurement. Other systems' figures are published JevBench leaderboard values as of September 2026. Noma's hard-tier figure covers the 111 public items; on a held-out half it scores 46.4%. Hosted latencies include a network round trip; Noma's was measured on the same machine as the server.
Pickbythekindofquestionyouask.
Your decisions are single-pass: routing, triage, intent, moderation, policy checks, and verifying agent steps. You want to self-host, you care about latency and cost, and you want open weights.
Your questions need multi-step reasoning, such as chained date arithmetic or long policies with amendments, and you prefer a managed API. Jev scores 74.1% on JevBench's hard tier against 51.4% for Noma.
Run Noma on every decision and send the ones it abstains on, or that you know need reasoning, to Jev or a reasoning model. The API is the same, so the switch is a URL.
# Switching an existing Jev client to Noma is a base-URL change
pip install blackdrome-noma
noma serve --port 8000
curl -s http://127.0.0.1:8000/v1/systemone \
-H "Content-Type: application/json" \
-d '{"state": "Deploy finished. 3 of 12 pods failing readiness.",
"questions": {"ok": {"type": "noul",
"instructions": "Did the rollout succeed?"}}}'Noma accepts the same request and returns the same response shape as Jev. Its two extra signals, abstain and uncertainty, sit under a separate key that existing clients ignore.
The full method, per-family results and the multi-step reasoning tier are in the evaluation document.
NomaandJev,answeredplainly.
- Is Noma an open-source alternative to Jev?
- Yes. Noma is an open-weight decision model released under MPL-2.0 that does the same job as Jev: it reads a state and typed questions and returns a probability for every option without generating text. You download the weights and run it on your own GPU, so there is no per-call fee and no data leaves your infrastructure.
- Does Noma work with existing Jev clients?
- Yes. Noma serves the same /v1/systemone wire format as Jev, so a client written for Jev works by changing the base URL. Noma adds two fields of its own, an abstain probability and an uncertainty estimate, under a separate key that existing clients can ignore.
- How fast is Noma compared with Jev?
- In our measurements with the JevBench client, Noma answers in 16 ms at the median, end to end over HTTP on one H100, with one question per request. The JevBench leaderboard lists Jev 1.13.0 at 652 ms. Jev's figure includes an internet round trip to a hosted API, while Noma's was measured on the same machine as the server.
- How accurate is Noma compared with Jev?
- On everyday single-pass decisions the two are close: Noma scores 100% on JevBench easy and 98.6% on the original tier, against 100% and 99.0% for Jev 1.13.0. On the hard tier, which needs multi-step reasoning, Jev is clearly stronger: 74.1% against 51.4% for Noma. If your questions need chained arithmetic or date reasoning, use Jev or a reasoning model for those.
- What does it cost to run Noma?
- By JevBench's method, which multiplies input tokens by the hosted price for the model's size class, Noma costs $0.023 per 1,000 decisions, against $0.040 for Jev 1.13.0. Self-hosted on the H100 we measured on, one serial stream costs about $0.026 per 1,000 decisions. The weights are free.
- What hardware does Noma need?
- An NVIDIA GPU with 6 GB of memory or more. The weights are one 5.4 GB file and use about 5.2 GB of GPU memory. The published latencies are from a Linux machine with an H100 and the fast path enabled; it also runs on Apple silicon and CPU, more slowly.
- How do I install Noma?
- Run pip install blackdrome-noma, then noma serve. That downloads the weights from Hugging Face, starts the API on /v1/systemone, and opens a local playground.
- What is Noma not good at?
- Multi-step reasoning. Noma answers in a single forward pass with no scratchpad, so questions that chain several calculation steps, or trace a long policy through its amendments, are outside its scope. It is also text only, limited to 4,096-token states, and evaluated in English.
Jev is a product of TypeSafe. Blackdrome AI Labs is not affiliated with TypeSafe. Product names are used only to describe compatibility and to compare published results.