codiv
CodivLiveOpenJev 0.1

Inference for openSystem One models

Typed answers with calibrated confidence, in milliseconds. Codiv serves OpenJev through a Jev-compatible API. Every account gets 100M input tokens free.

Get a free API keyTry the playgroundRead the docs

See it in action

codiv
$

Decisions, not tokens

A System One model reads your data once and answers every question with a probability, like a fast intuition you can measure.

Typed answers

Ask yes/no, choice and score questions. Every answer fits your schema by construction, because nothing is generated or parsed.

Calibrated confidence

Each answer is a full probability distribution with a confidence score, so thresholds actually mean something.

Milliseconds

One denoising step answers a whole set of questions. A typical request takes tens of milliseconds of model time.

Jev-compatible

The same wire API as TypeSafe's Jev. Their Python and TypeScript SDKs work unchanged; change the base URL.

Open all the way down

Open weights and an Apache-2.0 server. Anything you build on Codiv, you can also run on your own GPU.

Free to start

100M input tokens for every account while the experiment runs. No card required.

Playground

Try it in the browser

Typed questions, photos, chess, gomoku and chat, with every probability on screen. Sign in and run them on your free quota.

Any state and questions, with every request option.

Try it

state

Hi, I think I was charged twice for September. Could you take a look when you get a chance?

questions

  • urgentnoulNeeds a reply within the hour?
  • teamchoiceoutage · billing · feature
  • tonescorecalm · annoyed · furious
urgentno
yes
1%
no
99%
teambillingconf 0.94
billing
99%
outage
1%
feature
<1%
tonecalmconf 0.67
calm
90%
annoyed
9%
furious
1%

one denoising step · 38 ms

A replay of typical answers. The playground runs the real API on your free quota.

Models

Open models, served fast

OpenJev 0.1

Live

DiffusionGemma 26B-A4B turned into a System One model: the answer template goes onto the diffusion canvas and one denoising step reads a distribution for every answer.

Model id
openjev-latest
Question types
noul · choice · score
Base model
DiffusionGemma 26B-A4B, NVFP4
Modified by
mmastrac / razorback16
License
Apache-2.0, weights and server
Model detailsSource

Laya 1.0

Live

by Nandakishor M / Convai Innovations

A 421M ModernBERT-large encoder, fine-tuned on typed-decision workflows. Text only, about 10 ms per request.

Model id laya-1.0

Verdict 1.4

Live

by Heman10x

A 151M ModernBERT encoder with a GLiClass head, calibrated per option count. Text only, up to 24 choices, about 7 ms per request.

Model id verdict-1.4

CLM 0.1

Live

by Contrastive-LM

Contrastive state and action heads over Qwen3-8B embeddings: every option is scored against the state. Text only, 2,048 tokens, about 100 ms per request.

Model id clm-v0.1

JevK5 0.2

Live

by Alibi Serikbay

Qwen3.5-4B with a LoRA distilled from a thinking teacher, read from its answer letters' logits. Text only, 16,384 tokens, about 80 ms per question.

Model id jevk5-0.2

Drop-in

Already using Jev? Change one line.

Codiv speaks the same API as TypeSafe's Jev, down to the error shapes and request ids. Point the official SDKs at Codiv and your code keeps working.

Compatibility notes →
# Python or TypeScript: the official TypeSafe SDKs
export TYPESAFE_BASE_URL=https://api.codiv.ai
export TYPESAFE_API_KEY=sk-codiv-…

Free public experiment

Free for everyone

Free

Available now

$0

No card required

  • 100M System One tokens per account
  • 10M text-generation tokens
  • All models, including OpenJev
  • 60 requests per minute per key
  • Usage dashboard and up to 10 keys
Get a free API key

Roadmap

  1. OpenJev 0.1 public experiment

    Live

    DiffusionGemma 26B-A4B as a System One model, served free on api.codiv.ai.

  2. Dashboard, keys & usage

    Live

    Sign in with Google, GitHub or email. Track tokens per day.

  3. Image input

    Live

    Ask typed questions about up to 8 images per request, alongside the state.

  4. Text generation

    Live

    The same DiffusionGemma at an OpenAI-compatible /v1/chat/completions, with its own free quota.

  5. Playground

    Live

    Try reads, photos, chess, gomoku and chat in the browser, with every probability shown.

  6. Laya and Verdict

    Live

    Two small, fast open System One models from other authors, on the same API: laya-1.0 and verdict-1.4.

  7. CLM and JevK5

    Live

    Two more open System One models from other authors, served from a second GPU box: clm-v0.1 and jevk5-0.2.

  8. OpenJev 1.0

    Planned

    A new model of our own, built to raise the bar for fast, calibrated decisions. We think it will surprise you. Coming soon.

  9. Batch mode

    Planned

    Send thousands of states in one call and collect the answers when they're ready. Built for backfills and nightly jobs.

  10. More open System One models

    Planned

    More open models and sizes, from us and from the community.

Make your first decision in five minutes

Sign up, create a key and call the API. It's free.

Get a free API keyQuickstart