AnyJev vs localjev
Nokia research: turn any open LLM into a Jev-style decision model. Typed choice, yes/no or score from one prefill, de-biased with no labels or a fitted head, served on vLLM. — versus — GitHub Next's local Jev bridge: a Bun/TypeScript POST /v1/systemone that translates typed decision questions into prompts for DiffusionGemma behind any OpenAI-compatible endpoint.
Both give an existing model a Jev-style decision interface; localjev prompts DiffusionGemma through an HTTP bridge, AnyJev reads calibrated probabilities from the model's own next-token distribution.
| AnyJev | localjev | |
|---|---|---|
| Stars | 998 | 792 |
| Forks | 130 | 53 |
| Language | Python | TypeScript |
| License | Apache-2.0 | MIT |
| Last activity | 3 days ago | 13 days ago |
| Topics | decision-models, local | decision-models, gateway, local |
| Curated connections | 4 | 7 |
AnyJev — the curator's take
The cleanest way to get decisions out of a model you already run: ask a typed question, read the next-token distribution from one prefill, nothing generated or parsed. Level L0 removes option-position bias with zero labels; L2 fits a tiny closed-form head from 100 to 300 labels and serves it from a vLLM pooling endpoint, and `anyjev.pipeline` measures accuracy, calibration and latency on your own box instead of asking you to trust their tables. One finding worth stealing: cutting Qwen2.5-7B to 18 of 28 blocks was faster with slightly better accuracy. It is a week-old research repo from two authors, so expect API churn, and it assumes you can run an open model; without a GPU, laya's small checkpoints are the lighter path.
localjev — the curator's take
Use LocalJev when you want to point TypeSafe's SDK at your own hardware today, with a model you already serve: set TYPESAFE_BASE_URL and the quickstart runs unchanged, with chunking, admission control (max-inflight, HTTP 529 queue), and corrective retries on malformed JSON already handled. Read the honesty in its own README before trusting it: the probabilities are *generated* by the model as a JSON scalar/vector, not read from logits, so it is wire-compatible with Jev but not mathematically equivalent — OpenJev's structured read needs unmerged vLLM extensions that oMLX doesn't expose. That makes it fine for routing and triage, and the wrong tool for consequential decisions until you have run its own bake-off (AG News / BoolQ / SST-5, five models, two input lengths) on your workload. If you would rather own calibrated probabilities than borrow them, kev trains a readout head instead of prompting.