--- title: "Open replicas and Jev-compatible servers" type: community source_tier: community tags: [community, replicas, open-models, local-serving] created: 2026-09-25 updated: 2026-09-25 confidence: medium sources: - raw/x-repos/fstandhartinger__jevbench__results-v1-4-2.md - raw/x-repos/fstandhartinger__jevbench__docs-method-v1-4.md - raw/x-repos/fstandhartinger__jevbench__docs-release-v1-4-2.md - raw/community/huggingface-co-datasets-localllama-typed-decisions.md - raw/community/huggingface-co-datasets-sargedev-jev-distill-corpus-v3.md - raw/community/classifier-dev-about.md - raw/community/classifier-dev-pricing.md - raw/community/classifier-dev-terms.md - raw/x-repos/chigwell__typesafe.pro.md - raw/x-repos/chigwell__typesafe.pro__docs-admission.md - raw/x-repos/andrelandgraf__safer-with-jev.md - raw/x-repos/andrelandgraf__safer-with-jev__design.md - raw/x-repos/jaredpalmer__kev.md - raw/x-repos/Mapika__decider.md - raw/x-repos/kshetrajna12__reflex.md - raw/x-repos/kshetrajna12__reflex__docs-results-jevbench-official.md - raw/x-repos/kshetrajna12__reflex__docs-results-readme.md - raw/x-repos/bespokelabsai__nimble.md - raw/x-repos/bespokelabsai__nimble__docs-public-benchmarks.md - raw/x-repos/allebee__jevk5.md - raw/x-repos/allebee__jevk5__results-public231-jevk5-v0-2-summary-json.md - raw/x-repos/weitianxin__JevAny.md - raw/x-repos/weitianxin__JevAny__results-release-v0-2-json.md - raw/x-repos/abhishek085__open-spark-jev.md - raw/x-repos/abhishek085__open-spark-jev__model-card.md - raw/x-repos/abhishek085__open-spark-jev__docs-benchmarks.md - raw/x/CompleteSkeptic-2099925682726002904.md - raw/x-repos/iapp-technology__openthai-systemone.md - raw/x-repos/iapp-technology__openthai-systemone__docs-model-card.md - raw/x-repos/malevrigns__agent-jev.md - raw/x-repos/malevrigns__agent-jev__typed-decisions-comparison-json.md - raw/x-repos/zwliJay__jev-forge.md - raw/x-repos/zwliJay__jev-forge__docs-rlcd-results.md - raw/x-repos/intikhab49__open-jev-typed-decision-engine.md - raw/x-repos/Contrastive-LM__CLM.md - raw/x-repos/Contrastive-LM__CLM__examples-t-rex-results-jev-realtime-json.md - raw/x-repos/Contrastive-LM__CLM__examples-t-rex-results-clm-realtime-json.md - raw/x-repos/hunkim__solar-mini4-jev.md - raw/x-repos/pst2154__Nemotron_Jev.md - raw/x-repos/tinnel123666888__OmniJev.md - raw/x-repos/tinnel123666888__OmniJev__docs-scorecard-v05.md - raw/x-repos/OmniJev__PlayJev.md - raw/x-repos/kikoncuo__jevfire.md - raw/x-repos/wfzyx__von.md - raw/x-repos/Heman10x-NGU__openJev-verdict-2.0.md - raw/x-repos/andrueandersoncs__visual-jev.md - raw/x-repos/liushiliushi__JevTuner.md - raw/x-repos/9pings__notjev.md - raw/x-repos/9pings__notjev__docs-measurements.md - raw/x-repos/fidecastro__jevify.md - raw/x-repos/githubnext__localjev.md - raw/x-repos/githubnext__localjev__eval-reports-2026-09-18-bakeoff-report.md - raw/x-repos/razorback16__openjev.md - raw/x-repos/logan-markewich__jeff.md - raw/x-repos/logan-markewich__jeff__bench-results.md - raw/x-repos/chengyongru__fastjev.md - raw/x-repos/chengyongru__fastjev__docs-results.md - raw/x-repos/TheoLeeCJ__SemIf.md - raw/x-repos/deepanwadhwa__OpenDecision.md - raw/x-repos/ikermoel__open-alternative-jev.md - raw/x-repos/ikermoel__open-alternative-jev__benchmarks-docs-results.md - raw/x-repos/NotXf1le__choosekit.md - raw/x-repos/NotXf1le__choosekit__benchmarks-supergpqa-benchmark-json.md - raw/x-repos/zhangcy122__OpenJev.md - raw/x-repos/zhangcy122__OpenJev__examples-benchmark-results-json.md - raw/x-repos/Rizzo-AI-Academy__rizzo-flow.md - raw/x-repos/Rizzo-AI-Academy__rizzo-flow__results-readme.md - raw/x-repos/taeold__djev-run.md - raw/x-repos/emnlmn__snap.md - raw/x-repos/yoheinakajima__glance.md - raw/x-repos/yoheinakajima__glance__site-vslaya-index-html.md - raw/x-repos/HITsz-TMG__JevEmbed.md - raw/x-repos/HITsz-TMG__JevEmbed__reports-jevbench-public.md - raw/x-repos/HITsz-TMG__JevEmbed__reports-jevembed-data-test.md - raw/x-repos/nokia-applied-research__AnyJev.md - raw/x-repos/featherless-ai__simple-jev.md - raw/x-repos/yijunyu__jev-rs.md - raw/x-repos/ziozzang__hearim.md - raw/x/theanandprasad-2102199067057307900.md - raw/x-repos/NandhaKishorM__laya.md - raw/x-repos/receptron__laya.md - raw/x-repos/0xBakeer__arbiter.md - raw/x-repos/0xBakeer__arbiter__showcase-readme.md - raw/x-repos/jlt-commons__lev.md - raw/x-repos/jlt-commons__lev__bench-readme.md - raw/x-repos/bladedevoff__stuntd.md - docs/sweep/2026-09-25-mrjev/parta.md jev_version: "jev-1.13.0" summary: "Open models and servers that copy Jev's /v1/systemone shape but are not Jev: which is which, their own numbers (JevBench by version), warnings on keys, look-alike names, distillation." --- # Open replicas and Jev-compatible servers > **TL;DR** None of these is TypeSafe or runs Jev. They copy the wire shape of `/v1/systemone`, not Jev's calibration, and most compute `confidence` their own way, so thresholds tuned on Jev do not transfer. On the one independent board with a sealed set, JevBench v1.4.2 (2026-09-24, 93 systems), Jev 1.13.0 is #2 at 63.29 behind decider-4b v2 at 64.13, and leads on reasoning and calibration; the board weighs speed and cost equally with those. Pick a replica only for offline, private or cheaper runs, and re-measure on your own labels ([[ideas/measurements]]). Read the Warnings first. Split from [[ideas/sdks-and-replicas]] on 2026-09-25 (nothing dropped). Numbers are each author's unless marked. ★ · push as captured 2026-09-23 to 25. Pattern IDs: [[ideas/patterns]]. "Our check" = our 2026-09-25 source and registry vetting (docs/sweep record), not a raw capture. ## Quick picks by need From the rows below; none is Jev. - **Highest on JevBench v1.4.2:** decider-4b v2 (#1, 64.13 vs Jev 63.29), but Jev leads reasoning and calibration there, decider leads speed and cost, its 2026-09-24 models failed their own pre-registered release rules, and "nothing distilled" is `unverified`. Next: JevK5 (#3, 62.04; self-disclosed MMLU-Pro test-item replay in v0.2), reflex 4B (#7, 53.99; MIT, ships a frozen model and publishes its negatives). - **Small or edge:** Laya (32.8 ms per question on a T4; Node port @receptron/laya), but its base checkpoints fall below the majority baseline on typed decisions and the English one stays confident on Khmer and Hindi it fails; openJev-verdict-2.0 (149.6M, WebGPU); Von (395M; its own table 72.0% vs Jev 96.6% on 49 tasks); OpenThai-SystemOne (0.8B; 74.3 vs Jev 76.0 on Nimble's suite). Any logprob endpoint: NotJev (Qwen3-8B Q4 p50 23 ms). - **Avoid:** SiliconLabAI/OpenJev (any web page you visit while it runs can take its keys), solar-mini4-jev when you need `confidence` (hard labels), ollaya and deepopen (look-alike, renamed copy), and servers that silently answer to `jev-latest` (check the response `model`). With razorback16/openjev, never put a TypeSafe key beside the Codiv base URL. Details: Warnings. - **Before any swap:** Jev-tuned thresholds do not transfer; re-measure on your own labels and test option order, irrelevant state, missing evidence and general judgement ("What the replicas' own negatives say"). ## Which is which | Name | Means | Not to be confused with | |---|---|---| | OpenJev (four) | [TheoLeeCJ/openjev](https://github.com/TheoLeeCJ/SemIf), now **SemIf** (Qwen3.5-4B logit reader; JevBench v1.4.2 #11, 47.69; row on [[ideas/repos-apps]]; since the 2026-09-19 capture it added CPU via llama.cpp and Torch, and a 27B EXL3 bridge, our check; MLX is in the capture) · [razorback16/openjev](https://github.com/razorback16/openjev) (DiffusionGemma server, hosted on Codiv) · [IamBusy/OpenJev-Vision](https://github.com/IamBusy/OpenJev-Vision) (image events) · Zefan Cai's **Open-Jev** (a dataset plus 2B and 9B models; v1.4.2 #76 and #75) | also zhangcy122/OpenJev (sold as OpenJevPro) and SiliconLabAI/OpenJev (Warnings). PyPI `openjev` is balys's, none of these (our check) | | jevify (three) | [fidecastro/jevify](https://github.com/fidecastro/jevify): a server, owns PyPI `jevify` | altryne/jevify ([[ideas/tools-and-integrations]]), ryana/jevify, a prompt ([[ideas/repos-agents]]) | | reflex | [kshetrajna12/reflex](https://github.com/kshetrajna12/reflex), whose server command is `reflex-serve` (the "Reflex" local server some tools name) | lateos-ai/reflex (a Rust/CUDA engine without `/v1/systemone`); PyPI `reflex` (a web framework) | | djev (three) | Davipar/djev-dev with api.djev.dev (Maisa; the one JevBench ranks, v1.4.2 #8) | mmastrac/djev; [taeold/djev-run](https://github.com/taeold/djev-run), built on mmastrac's | | AnyJev / JevAny | [nokia-applied-research/AnyJev](https://github.com/nokia-applied-research/AnyJev) | [weitianxin/JevAny](https://github.com/weitianxin/JevAny) | | OmniJev | [OmniJev/PlayJev](https://github.com/OmniJev/PlayJev) (games) | [tinnel123666888/OmniJev](https://github.com/tinnel123666888/OmniJev) (multimodal); Iron-LYK/OmniJev returned 404 (our check) | | Laya | [NandhaKishorM/laya](https://github.com/NandhaKishorM/laya) = HF `convaiinnovations/laya` | HF `convaiinnovations/laya-typed-decisions` is a separate checkpoint fine-tuned on that benchmark's train split | **`confidence` is not one thing.** Jev's field is its own ([[concepts/confidence]]). Replicas use normalised entropy `1 − H(p)/ln K` (notjev, razorback16/openjev, JevEmbed; localjev and hearim say entropy-based), the top probability (Nemotron_Jev), a "peak" formula (jevify, OmniJev, ollaya, JevAny, our check) or nothing real (solar-mini4-jev answers one letter, so every answer is certain). Same field name, different number. **Replicas that answer to Jev's model names.** localjev, razorback16/openjev, jeff and arbiter accept `jev-latest` (localjev and openjev also `jev-preview`). An SDK whose base URL points at one gets local answers with no error; only the response `model` field tells. fastjev rejects `jev-latest`. ## Warnings Plain names on purpose: no links. Flagged, not adjudicated; our reading of the MCA ([[reference/legal-and-data]]), not legal advice. - **(a) Public Jev on the operator's key** (possible MCA §2.3(a): offering the Services as a standalone service). **typesafe.pro**: a keyless front door with free and paid token tiers that pools several operator master keys (`TYPESAFE_MASTER_API_TOKEN_N`, 1,200 requests/minute and 250,000 tokens/second each, per its docs); spreading load over several keys may also meet §2.3(j), Usage Limits (inferred). **classifier.dev**: free with no key (20,000 requests/day per IP), paid Pro tier; says its fast tier is Jev; its /about page names the operator as Michael Ryaboy, an independent developer; its terms mention partner keys, and whether it holds a TypeSafe agreement is unknown. **safer-with-jev**: a public demo API with no caller auth that runs anyone's text through Jev on the owner's key (1,000 Jev calls/day deployment-wide). Warnings, never routes. - **(b) Distillation** (MCA §2.3(b) forbids training a model to imitate the Services' output). **SargeDev jev-distill-corpus-v3**: its own card says its largest stream, 498,010 of 740,957 rows, has labels distilled from Jev 1.13 via OpenRouter. **JevEmbed-Data** includes about 38k rows of it (our check of the dataset). **laya-jev-GraphRAG** pitches paired Laya/Jev answers as training data for Laya (our check). **Bespoke Nimble** ships Jev-labelling scripts and says saved Jev probabilities are "available for future soft-target distillation"; the released training set looks clean (our check), and its "did not distill from Jev" is `unverified`. **stuntd**'s paid-upstream mode learns from TypeSafe's answers and then takes the decision over. - **(c) Key safety.** **SiliconLabAI/OpenJev** attaches the server's TypeSafe, decider or OpenAI key to a URL taken from the request body, with CORS open to all origins: any web page you visit while it runs can take those keys (our check; not listed). **razorback16/openjev**'s instructions put a Codiv key in `TYPESAFE_API_KEY` beside `TYPESAFE_BASE_URL=https://api.codiv.ai`; a builder who changes only the base URL sends a real TypeSafe key to Codiv (inferred). Replicas that answer to `jev-latest` (above) swap the model silently. - **(d) Install names that don't exist or belong to someone else** (our registry check): PyPI `open-alternative-jev` and `deepopen` do not exist; PyPI `llm2jev` is tic-top's, not Yinsongxu's; PyPI `openjev` is balys's; npm `fastjev` is an unrelated squat (use PyPI `fastjev`); PyPI `reflex`, `glance`, `snap` are unrelated. On tools pages: npm `jevwire` is 404, npm `pi-jev-guard` and `pi-jev-router` belong to other publishers ([[ideas/tools-and-integrations]]). ## Models with their own weights | Tool (owner) | What | Install | ★ · push | Their numbers and caveats | |---|---|---|---|---| | [Kev](https://github.com/jaredpalmer/kev) (Jared Palmer) | 0.8B/4B/9B/27B on Qwen bases, plus the original Kev-0.5B (Qwen2.5-0.5B) kept for reference; serves `/v1/systemone` (+ `/v1/systemone/permute`, 1–64 option orders) | HF weights, `uv sync` | 6875 · 2026-09-25 · Apache-2.0 | README of 2026-09-25: Kev-27B 0.848 vs Jev 0.857 on new sources; TypeSafe's 102 public eval questions, agreement Kev-9B 0.809 vs Jev 0.891; MMLU Kev-9B 0.74 vs Jev 0.90, MMLU-Pro 0.52 vs 0.84; trained on states ≤ 384 tokens, serves 8,192 per state and per question, accuracy drops on long documents. JevBench v1.4.2: kev 4B #28 (36.14), 8B #49, 0.6B #53 (these three marked research preview), 0.5B #59 (18.88). Version-dependent: quote the model and date. "No Jev outputs were used for training": `unverified` | | [decider](https://github.com/Mapika/decider) (Mapika) | Qwen3.5 fine-tunes (2B, 4B, 35B-A3B), local Qwen3.5-27B teacher | `pip install decider-ai` | 394 · 2026-09-25 · Apache-2.0 | README: public hard tier Jev 0.730 vs decider-35b-a3b 0.676; publishes that its 2026-09-24 models failed their pre-registered release rules. **JevBench v1.4.2: decider-4b v2 #1 at 64.13** vs Jev 63.29; Jev leads Intelligence (53.1 vs 49.4) and calibration, decider leads speed and cost (board's note); decider-35b-a3b #19, decider-2b #39. "Nothing was distilled from Jev": `unverified` | | [reflex](https://github.com/kshetrajna12/reflex) (Kshetrajna Raghavan) | frozen Qwen3.5-4B (default) or Qwen3.8-27B read by option letter, two option orders averaged; no weights of its own; `reflex-serve` | `git checkout stable && uv run reflex-serve --stable` | 147 · 2026-09-23 · MIT | By version: JevBench v1.2 run by the benchmark author (534 decisions, 2026-09-21): reflex 4B (a LoRA configuration, not today's `stable`) #5 at 71.7 vs Jev #1 75.4; reflex-27b hard tier 75.9% vs Jev 74.1%, top calibration axis 86.2, but #22 on cost and speed. v1.4.2: reflex 4B #7 (53.99), reflex-27b #64. Self-run public items: hard 0.685 / ECE 0.081 vs Jev 0.730 / 0.031. Publishes that every fine-tune and GEPA prompt tuning won in-distribution and lost general judgement, so it ships the frozen model | | [Bespoke Nimble](https://github.com/bespokelabsai/nimble) (Bespoke Labs) | Nimble-9B, LoRA on Qwen3.5-9B; data from GPT-5.6 | HF `bespokelabs/Bespoke-Nimble-9B` | 1795 · 2026-09-24 · no repo licence (weights Apache-2.0, our check) | **Independent suite, Jev run live on human labels** (13 public subsets, 3,880 records, contributed by Edgar Dyck): Jev 1.13.0 76.0% macro vs Nimble 74.8%; Jev leads Noul 84.6 vs 80.2 and Choice 82.9 vs 81.6, Nimble leads Score 54.6 vs 50.1; Jev 35.0% on SummEval-relevance; Jev lower ECE on 11 of 13; MASSIVE en → de Jev 87.4 → 86.9, Nimble 86.9 → 83.4 ([[ideas/head-to-head-benchmarks]]). JevBench v1.4.2 #60. Distillation note: Warnings (b) | | [JevK5](https://github.com/allebee/jevk5) (Alibi Serikbay) | Qwen3.5-4B + LoRA reading option letters (SemIf method, credited); also 9B, 2B, Lite, GGUF | HF `alibiserikbay/JevK5` | 107 · 2026-09-25 · Apache-2.0 | Author on JevBench public 231: hard 0.784 vs base 0.613; the 9B is worse than the 4B. **JevBench v1.4.2 #3, 62.04 vs Jev 63.29; sealed 308 items Jev 36.7% vs JevK5 33.1%.** Self-disclosed: v0.2 replayed MMLU-Pro test items. Teacher includes GPT-6 Luna (an OpenAI-terms question). "No Jev outputs": `unverified` | | [JevAny](https://github.com/weitianxin/JevAny) (Tianxin Wei) | 27B pointer-head model with an RLCR calibration reward; includes Kev code (NOTICE) | HF `tianxinwei/JevAny-27B-SFT` / `-RLCR` | 16 · 2026-09-25 · Apache-2.0 | Author: transfer accuracy SFT 82.41% vs Jev 85.37%; MMLU-Pro 73.0% vs Jev 84.0%. No per-item Jev file, so live vs copied is unknown. Own negatives: RLCR no gain; label-free self-training negative. "No Jev outputs": `unverified` | | [open-spark-jev](https://github.com/abhishek085/open-spark-jev) (Abhishek Rai) | LoRA of Qwen3.5-4B for DGX Spark (`spark-s1`), 11,792 rows from its own open-model data factory; `/v1/evaluate` with a `/v1/systemone` alias | HF `abhishek085/spark-s1-4b-v6` | 18 · 2026-09-22 · Apache-2.0 | Where Jev leads (author, Jev figures from third-party per-row records, not run live): 60 tool calls 0.900 vs Jev 0.917 (Jev p50 421.6 ms with network, local 74.9 ms); Kev transfer 0.756 vs 0.857; vulnerable code at chance (0.502). **RLCD made results worse twice; not shipped.** Its JevBench 83.1 is a self-computed proxy; v1.4.2 official #15, 44.62. Its "20–200x" is the CEO's launch-post range ([post](https://x.com/CompleteSkeptic/status/2099925682726002904)), not the blog's 40x–200x | | [OpenThai-SystemOne](https://github.com/iapp-technology/openthai-systemone) (iApp Technology) | Qwen3.5-0.8B with a 256-way slot head, Thai continued pretraining, synthetic data from Qwen3.6-35B-A3B | `pip install openthai-systemone`; hosted API (our check) | 60 · 2026-09-21 · Apache-2.0 | On Nimble's 13-subset suite: 74.3 vs Jev 76.0 (Jev's copied from Bespoke). Weak spots it states: SummEval-relevance 21.7, PubMedQA 64, English ECE ~0.15; v0.3 trained on train splits of 5 bench sets; option-order fix cuts flips 18.9% → 3.3% on its Thai set | | [AgentJev](https://github.com/malevrigns/agent-jev) (malevrigns) | Qwen3-0.6B with a candidate head; own `POST /api/evaluate`, not `/v1/systemone` (no SDK route) | HF `aimeigaoshou/agent-jev` (a different account) | 303 · 2026-09-23 · Apache-2.0 | typed-decisions test: 79.25% vs a local Laya run 77.00%; Jev's 72.7% is quoted from the typed-decisions card, and its own README says the rows are not comparable (specialists trained on the benchmark). Its Claude Code hook fails open | | [JevForge](https://github.com/zwliJay/jev-forge) (zwliJay) | next-click scoring on website-disjoint Mind2Web splits; `/v1/systemone` without auth | HF `AndeyTait/JevForge-0.8B` | 53 · 2026-09-23 · bare MIT line | Author: Choice top-1 test/OOD 0.579 / 0.637 vs Jev 0.543 / 0.610; **Jev wins Noul** (0.910, Brier 0.092 vs 0.826 / 0.128) and score MAE. Its Jev benchmark caps at 200 records per split and drops failed calls (our check), so not like-for-like. Label source unnamed. P12 | | [open-jev-typed-decision-engine](https://github.com/intikhab49/open-jev-typed-decision-engine) (intikhab49) | ModernBERT-base 150M, labels in the input; trained on typed-decisions train (1,016 cases) | clone | 44 · 2026-09-21 · Apache-2.0 | Test: fine-tune 0.624; ensemble 0.6965 / ECE 0.0565, only for known question sets; unseen questions ≈ 0.624. Jev 0.727 quoted. Its "single human annotator 0.659" misreads the card, whose gold is a model teacher | | [CLM](https://github.com/Contrastive-LM/CLM) (Contrastive-LM) | frozen Qwen3-8B embedder + 20M two-tower head; serves `/v1/systemone`; calls Jev only as a baseline | `pip install contrastive-lm` | 1187 · 2026-09-24 · Apache-2.0 | T-Rex, 5 seeds × 60 s, safety shield on for both: both survive 5/5, but **Jev agreed with the physics planner 0.987 vs CLM 0.658**, so CLM's survival leans on the shield; p50 149.8 vs 16.5 ms. Its best-of-N verifier claims against Jev (DeepSWE, Terminal-Bench) have no Jev file in the repo: `unverified`. JevBench v1.4.2 #78 (8.57) | | [solar-mini4-jev](https://github.com/hunkim/solar-mini4-jev) (Sung Kim) | Upstage Solar chat model answers one option letter per question (BYOK Upstage key) | clone; serves `/v1/systemone` | 43 · 2026-09-24 · no licence | **Hard labels**: no usable confidence, so no confidence routing. Author, 886 public items: Jev 75.6 vs solar-pro4 76.9, solar-mini4 71.4; Jev tied only on classifier-benchmark v1 (97.4). Prompt tuned on public Jev failure cases (overfit risk) | | [Nemotron_Jev](https://github.com/pst2154/Nemotron_Jev) (Alex Steiner) | masked-token scoring on NVIDIA's Nemotron diffusion LMs via modified vLLM | clone (no `main` branch, our check) | 14 · 2026-09-25 · MIT repo, base model under NVIDIA's Nemotron licence | Author: 174/231 on JevBench public items with answer reordering. No auth, TLS or rate limit (author warns). Probabilities uncalibrated. LoRA labels from an OpenAI model (OpenAI terms, not MCA) | | [OmniJev](https://github.com/tinnel123666888/OmniJev) (tinnel123666888) | decisions over images, video, screens, robot scenes, spectrograms; no HTTP server | HF `tinnel123/OmniJev` | 70 · 2026-09-25 · NOASSERTION | Author, held out: LIBERO-10 0.807, Mind2Web 0.733, POPE 0.902; A800 294 ms per question. Institutional affiliation self-stated only. Jev is text-only ([[reference/models-and-pricing]]) | | [PlayJev](https://github.com/OmniJev/PlayJev) (OmniJev) | Qwen3.5-0.8B-Base playing ten browser games from raw pixels | HF | 36 · 2026-09-24 · Apache-2.0 | 43 ms per move on an H200; 0.57 of its game teachers (trained with DAgger) | | [JEVfire](https://github.com/kikoncuo/jevfire) (kikoncuo) | Jev-inspired parallel decisions on vLLM (CUDA) and WebGPU | clone | 67 · 2026-09-18 | 28 decisions in 497 ms, 10.3× faster than generating the same constrained JSON on the same 27B model; Mario 71 ms/action in a browser (Qwen3.5 0.8B, M4 Max); credits the RLCD checkpoint | | [Von](https://github.com/wfzyx/von) (wfzyx) | 395M encoder | `pip install von-sdk` / `npm install von-sdk` (plain `von` is unrelated) | 526 · 2026-09-23 · Apache-2.0 | own table: Jev 96.6% vs Von 72.0% on 49 tasks. JevBench v1.4.2 #46 | | [openJev-verdict-2.0](https://github.com/Heman10x-NGU/openJev-verdict-2.0) (Heman10x) | 149.6M ModernBERT, confidence head, WebGPU | Hugging Face | 273 · 2026-09-20 · NOASSERTION | claims 77.10% vs Jev 72.70%; Jev's figure is from the typed-decisions card, where the dataset authors ran Jev live on 2026-09-18, and gold is agreement with a ~4B teacher (ceiling 0.735). JevBench v1.4.2: Verdict 1.4 #58, Verdict #62 | | [visual-jev](https://github.com/andrueandersoncs/visual-jev) | image decisions, Qwen3-VL | uv | 2 · 2026-09-20 · no licence | Jev is text-only. Not hr98w/jev-visual (a teaching repo) | | [JevTuner](https://github.com/liushiliushi/JevTuner) | Brier-loss training recipe | clone | 12 · 2026-09-21 · no licence | not TypeSafe's method | ## Servers and readers over other models | Tool (owner) | What | Install | ★ · push | Their numbers and caveats | |---|---|---|---|---| | [jeff](https://github.com/logan-markewich/jeff) (Logan Markewich) | `/v1/systemone` over GLiFormer (knowledgator, 400M); accepts `jev-latest`; 529 on a full queue, 429 with `retry-after-ms` | clone | 246 · 2026-09-20 · MIT | **Jev run live**, 1,600 labelled items: AG News 75.5% vs Jev 90.5%; p50 151 vs 129 ms. JevBench v1.2.2: jeff 66.9 (#9 of 18) vs Jev 75.3 (#2); hard tier 37.7% vs 74.1%. v1.4.2: #40, 30.58. Self-run hard items: Jev lost temporal/numeric 4/15 vs 6/15 (matches [[concepts/jaggedness-jev-1-13]]), led ambiguous 6/7 vs 0/7 | | [localjev](https://github.com/githubnext/localjev) (GitHub Next) | Bun bridge: any OpenAI-compatible local model writes probabilities as JSON (self-reported, not logits); accepts `jev-latest`, `jev-preview`; 529 above 64 queued | clone | 775 · 2026-09-18 · MIT | M5 Max bake-off, 120 items: short-input macro Qwen3.6-35B-A3B 76.7%. **2,048 unrelated words lowered every model's macro accuracy** (Qwen 76.7% → 69.2%; Gemma E4B SST-5 50.0% → 12.5%). Evidence that prompted-probability replicas degrade under irrelevant state | | [OpenJev](https://github.com/razorback16/openjev) (razorback16) | DiffusionGemma 26B-A4B via patched vLLM or MLX, read as a diffusion canvas; also serves Laya and Verdict; extensions incl. images; accepts `jev-latest` | clone; hosted free on Codiv (100M input tokens) | 421 · 2026-09-24 · Apache-2.0 | Author, one GPU: p50 27 ms (1 question), 31 ms (3); 57.4 req/s at concurrency 64 (p95 1,109 ms). JevBench v1.4.2 #27 (36.85; thinking mode #68). **Codiv key routing: Warnings (c)** | | [NotJev](https://github.com/9pings/notjev) (Nathanael Braun) | letters the options, asks any logprob endpoint for one token (26 options max); abstains below `theta` | `npm i notjev` | 21 · 2026-09-22 · Apache-2.0 | Author: Qwen3-8B Q4 p50 23 ms; 27B NVFP4 on vLLM 101 ms; own judge 0.947 agreement on n = 1,224. Its "hosted Jev 419 ms" has no source in the repo. In `serve` mode one key is both upstream bearer and caller guard | | [jevify](https://github.com/fidecastro/jevify) (Felipe Infante de Castro) | any LLM as a Jev-like endpoint: hashed YAML recipe, state sent once, logprob ladder; endpoint, embedding, NLI or rerank backends | `uv tool install jevify` (PyPI) | 43 · 2026-09-23 · MIT | Unmodified `typesafe-sdk` works against it (author). policy-hard-52: DeepSeek-V4-Flash 47/52, Gemma 4 E4B 45/52. No Jev comparison | | [fastjev](https://github.com/chengyongru/fastjev) (chengyongru) | SDK-first SemIf fork: Torch, vLLM, MLX, llama.cpp GGUF, EXL3; 2–16 options; rejects `jev-latest` | `pip install 'fastjev[torch]'` (npm `fastjev` is unrelated) | 21 · 2026-09-24 · MIT | RTX 5090: vLLM batched 36.26 vs Torch 18.52 decisions/s. Jev side copied (SemIf's 0.845 vs Jev 0.883 on 102 rows; count once). Headline says "open source implementation" while its results say Jev is not reproduced | | [OpenDecision](https://github.com/deepanwadhwa/OpenDecision) (Deepan Wadhwa) | zero-shot NLI (ModernBERT-large) behind the Jev shape, plus retrieval | `pip install OpenDecision` (Python ≥ 3.13) | 57 · 2026-09-21 · Apache-2.0 | Treat scores as uncalibrated (README). JevBench v1.4.2 #56, 21.64. P24 | | [open-alternative-jev](https://github.com/ikermoel/open-alternative-jev) (Iker Moel Tacher) | in-process library: all questions about one state in one pass, option-letter logits; not SDK-compatible | clone (the PyPI name in its README does not exist) | 54 · 2026-09-25 · Apache-2.0 | Stock Qwen3.6-27B 8-bit on typed-decisions: 73.7% / ECE 0.020 vs Jev 72.7% / 0.144 (Jev copied; gold is teacher agreement, ceiling 0.735). **Reversing options moved 4B yes/no accuracy 13.5 pts**; packing hurts below ~4B. v1.4.2 #36. P01 | | [choosekit](https://github.com/NotXf1le/choosekit) (Felix Koba) | TS: option-label logprobs from llama.cpp, Ollama or OpenRouter, with images; `choosekit-mcp` | `npm i choosekit` | 23 · 2026-09-24 · Apache-2.0 | **Jev run live** (OpenRouter): SuperGPQA 1,000 items, Jev 53.6% at $0.0244 per 1,000 decisions, p50 325 ms; Kimi K3 59.3% at ~25× the cost. Graduate knowledge questions are a poor fit ([[concepts/jaggedness-jev-1-13]]). SemIf authored144: tie at 139/144 with local Qwen3.8 27B Q4 | | [OpenJevPro](https://github.com/zhangcy122/OpenJev) (Corel Zhang) | letter-logprob scoring on Ollama or OpenAI-compatible models; "guard" wrapper around Jev | `pip install openjevpro` | 30 · 2026-09-25 · **PolyForm Noncommercial** + paid licence | Author, 60 Banking77 cases (`benchmark_results.json`): Jev 98.3% = an Ollama model 98.3% (`gpt-oss:20b-cloud` per its benchmark script, our check; the results file says only "ollama"), mean 732 vs 1,937 ms; separately, 36 queries: Jev 97.2% (35/36), p50 740 ms. Its "Jev ~25–45 ms TTFT" contradicts its own 732–750 ms; "60%+ cost cut" and "99.99% SLA" are unmeasured. Tiny n | | [rizzo-flow](https://github.com/Rizzo-AI-Academy/rizzo-flow) (Simone Rizzo) | option-letter logits from one shared prefill; adds `numeric` and an "insufficient evidence" option | clone | 459 · 2026-09-25 · Apache-2.0 | Author (RTX 5060 Ti): SemIf authored144 0.812 vs SemIf 0.819; p50 49 ms. **6 of 36 confident wrong answers when evidence is missing.** "No Jev labels": `unverified` | | [djev-run](https://github.com/taeold/djev-run) (Daniel Lee) | Cloud Run deployment of mmastrac/djev on DiffusionGemma | ghcr.io image | 551 · 2026-09-24 · no licence | Author: $3.19/h on one RTX PRO 6000, scales to zero; cold start ~47.5 s; 117 ms median. Its "JevBench 73.4" is self-run, not an official row. Answer shape differs from the docs (1-indexed Score probabilities, no `type` field; our check) | | [SNAP](https://github.com/emnlmn/snap) (Emanuele Menon) | Rust over vendored llama.cpp; adds `numeric`, `allow_abstain`, and a `confidence` on Noul (Jev has none) | build | 21 · 2026-09-25 · no licence file | Author: Qwen3.8-4B agrees 73.2% with frontier-consensus labels over 373 TypeSafe public decisions; **option-order reversal flips a third of MiniCPM's choices**; 51 ms vs Ollama 149 ms. Binary `snap` shadows Ubuntu's | | [Glance](https://github.com/yoheinakajima/glance) (Yohei Nakajima) | logits from a frozen Qwen3-VL-4B for image decisions; `/v1/decide` | `pip install glance-vlm` | 23 · 2026-09-24 · Apache-2.0 | Pre-registered fresh photos: yes/no 0.939 (541) vs Gemini 3.1 Flash-Lite 0.961; exact rating 0.669 vs 0.763. **vs Laya Vision on an M5: 0.886 vs 0.689** ([page](https://github.com/yoheinakajima/glance/blob/d458a27654ad9258977526395547c51e632fa5cd/site/vslaya/index.html)). P34 | | [JevEmbed](https://github.com/HITsz-TMG/JevEmbed) (HIT Shenzhen) | any embedding model → Choice/Score/Noul by cosine and temperature; normalised-entropy confidence | clone | 26 · 2026-09-25 · Apache-2.0 | Author, JevBench public 231: best Qwen3-Embedding-8B 58.44%; that report was **deleted from the default branch on 2026-09-25** (captured from commit 9cb60bb). Dataset licences mixed, incl. CC-BY-NC; distillation note: Warnings (b). P17 | | [AnyJev](https://github.com/nokia-applied-research/AnyJev) (Nokia Applied Research) | any HF LLM, debiased, optional head on 100–300 labels | `pip install "anyjev[hf]"` | 267 · 2026-09-23 · Apache-2.0 | order flips 0.230 → 0.073 (Qwen3-8B) | | [Simple Jev](https://github.com/featherless-ai/simple-jev) (Eugene Cheah) | logits server, free demo API | pip from repo | 491 · 2026-09-21 · Apache-2.0 | own contract. typed-decisions card: 0.716 (Jev 0.727). JevBench v1.4.2 Qwen3.8-27B #32 | | [jev-rs](https://github.com/yijunyu/jev-rs) (Yijun Yu) | Rust over llama-server; serve, MCP, eval | `cargo install jev-rs` | 5 · 2026-09-22 · Apache-2.0 | SDKs reach it via `TYPESAFE_BASE_URL` | | [hearim](https://github.com/ziozzang/hearim) (ziozzang) | Go gateway over Ollama, vLLM, SGLang | `go build` | 5 · 2026-09-22 · NOASSERTION | entropy-based `confidence` | | Recipe ([@theanandprasad](https://x.com/theanandprasad/status/2102199067057307900)) | softmax over yes/no logits | — | post 2026-09-22 | 21/22 on Llama 3.3 70B; raw logits flip with option order (AnyJev) | **What the replicas' own negatives say:** option order moves answers (open-alternative-jev, SNAP, AnyJev), irrelevant state hurts prompted replicas (localjev), missing evidence produces confident errors (rizzo-flow), and fine-tuning tends to trade general judgement for in-distribution wins (reflex, open-spark-jev). Test all four before swapping one in ([[guides/testing-and-evaluation]]). ## Laya and its servers Laya ([NandhaKishorM/laya](https://github.com/NandhaKishorM/laya), Nandakishor M; `pip install laya`; 18593★ · 2026-09-23 · Apache-2.0): three encoder checkpoints plus a router, `laya-serve`; 32.8 ms per question on a T4. Its Jev figures are quoted, not measured; Jev leads Banking77 0.870 vs 0.425. **Independent caveats:** zero-shot on typed decisions the base checkpoints score 0.362 (English) and 0.352 (multilingual), below the 0.461 majority baseline (Laya's model card, quoted by arbiter; the 0.766 belongs to the checkpoint fine-tuned on that benchmark); the English checkpoint scores 0.000 on Khmer at 0.952 confidence and 0.100 on 20-way Hindi intent (chance 0.050) while staying confident (arbiter); lev's encoders reach 61–67% on SemIf's authored144; Laya Vision 0.689 vs Glance 0.886. JevBench v1.4.2 #41, 30.25. | Tool (owner) | What | ★ · push | Notes | |---|---|---|---| | [@receptron/laya](https://github.com/receptron/laya) | Node/TypeScript ONNX Runtime port, `npm i @receptron/laya` | 466 · 2026-09-21 · MIT | matches Python to four decimal places; ~140 ms per 3 questions on Apple-silicon CPU; ~1.7 GB download; options within 192 tokens, state cut at 512 | | [arbiter](https://github.com/0xBakeer/arbiter) (Khaled Bakeer) | serving layer: router, micro-batching, CUDA graphs; accepts `jev-latest`; MCP server; Claude Code hook that **fails open** | 33 · 2026-09-24 · MIT | GB10 20.9 ms per question; the Hindi collapse above; six games vs baselines: beats random on only 3 | | [lev](https://github.com/jlt-commons/lev) (jlt-commons) | Clojure: Laya encoder first, escalates below a gate to Qwen3.5-4B | 19 · 2026-09-24 · Apache-2.0 | encoders 61–67% vs a MiniCPM5-2B thinker 95.1% on authored144; gate at 0.5 gives 92.4% at 270 ms per case. Its "24/64 encoder vs 60/64 hosted Jev" has no file. P06 | Names only (C, not captured; our check): laya-mps (Apple MPS; its own `/v1/decisions`, not a drop-in), sys1 (Rust server, entropy confidence), laya-server (1Panel; own keys and admin UI), **ollaya** (an Ollama look-alike in name, CLI and port, installed by `curl | sh` that adds a systemd service), **deepopen** (mostly a renamed copy of Laya presenting Laya's numbers as its own; PyPI `deepopen` does not exist), laya-jev-GraphRAG (its Jev client reads the wrong response fields and hides errors; distillation: Warnings (b)). ## Names only Not captured (C; our check, 2026-09-25): OpenJev-Vision (IamBusy; image events, not general VQA), JEV-CPU (a CPU port of SemIf, superseded by SemIf's own CPU backends), dev-0.4b (no licence file), Dohnuts (weights CC BY-NC-SA; states Jev leads), JevBERT / typed-decision-bert (NLI shell), jev_local (Japanese; weights CC-BY-SA), LLM2Jev (Yinsongxu; PyPI `llm2jev` is another project's), OpenSourceJev (sabeel111; JevBench v1.4.2 #21, 40.87), Valen (image and video; Jev is text-only), jevper (a `typesafe-sdk`-style client over OpenAI/Anthropic models; its "SDKs retry 409" `contradicts docs`, [[reference/rate-limits-and-errors]]), Lateos reflex (see Which is which). Earlier names only: [LitJev](https://github.com/zhengxuyu/litjev) (v1.4.2 #57), [jevmlx](https://github.com/bnsd55/jevmlx), [jev-visual](https://github.com/hr98w/jev-visual). Elsewhere: SemIf, NanoJev, jevlike, openjev-sglang, mini-jev → [[ideas/repos-apps]]. ## Reading the numbers - **JevBench changes by version.** v1.2 (reflex's record, 2026-09-21): Jev #1 at 75.4. v1.2.2 (jeff's record): classifier.dev fast tier #1 at 84.8, Jev #2 at 75.3. v1.4.2 (2026-09-24): 534 frozen plus 308 sealed decisions; Jev #2 at 63.29, sealed accuracy 36.7% (chance 0.293). Jev's row is flagged API: TypeSafe's endpoint received the sealed item text, without answers, as every API row did. classifier.dev (fast tier = Jev) is listed, not ranked. Always quote the version. - **typed-decisions** (LocalLLaMA): the dataset authors ran Jev live on 2026-09-18 (`jev-latest` → `jev-1.13.0`, 400 cases, 2,000 decisions): 0.727, p50 710 ms. Gold is the mean of three samples from a ~4B-class teacher, so the score is agreement with that teacher (ceiling 0.735), not correctness. Replicas quoting Jev 0.727 copy this row. ## Related - [[ideas/sdks-and-replicas]] — community SDKs and libraries that call the real Jev - [[ideas/measurements]] — replica trust notes; [[ideas/head-to-head-benchmarks]] — JevBench and public datasets - [[ideas/platforms-and-gateways]] — hosted routes to the real Jev, and the same warnings - [[reference/legal-and-data]] — MCA §2.3; [[reference/environment-variables]] — `TYPESAFE_BASE_URL` ## Sources Links inline; raw captures (2026-09-23 to 25) in frontmatter. "Our check" items come from the 2026-09-25 vetting record (docs/sweep/2026-09-25-mrjev), not raw captures.