datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
jev-bench
jev-bench
Real human-labeled data, reformatted into System One questions — with human label distributions wherever they exist.
22 configs · 166,054 rows · 22,773 test records · 4 calibration-gold configs · v0.1.1
Repo & engine · Source rationale · What we verified about Jev's API · Other independent Jev evaluations
jev-1.13.0 on every test record: crisp, grounded decisions land in the accurate-and-calibrated corner; ordinal ratings and anything humans disagree about do not.… See the full description on the dataset page: https://huggingface.co/datasets/Praveenrajus/jev-bench.jev-stage2-image-beans-pilot
Beans: one natural question per image
Open the corrected preview.
natural_v4 is the recommended and default preview: 100 original images, 100 rows, one three-way condition-class Choice question per image. All targets come directly from the source labels column (34 angular leaf spot, 33 bean rust, 33 healthy). Original image bytes and source annotations are unchanged.
Example question: “Which source-defined condition class describes the bean leaf?” Options: angular_leaf_spot… See the full description on the dataset page: https://huggingface.co/datasets/FaroukMoc2/jev-stage2-image-beans-pilot.jev-luna-pagerduty-trigger
Jev vs Luna as a PagerDuty trigger
Synthetic checkout/payments log stream with gold labels from PagerDuty alerting principles: page only if a human must act now. TypeSafe’s Jev (typesafe-ai/jev) and GPT-5.6 Luna (openai/gpt-5.6-luna) both ran on Vercel AI Gateway. There is no ERROR auto-page.
This is not production traffic and not the Loghub junk-filter benchmark.
Write-up: https://github.com/reachjalil/jevlogs/blob/hf-benchmark/docs/article/jev-vs-luna-pagerduty.md
Code:… See the full description on the dataset page: https://huggingface.co/datasets/reachjalil/jev-luna-pagerduty-trigger.jevlogs-log-triage-benchmark
Jev Logs log-triage benchmark
A labeled evaluation of Jev Logs on sanitized public logs. Jev Logs asks TypeSafe’s Jev, through Vercel AI Gateway, whether a log line is worth sending to an expensive reasoning model. This dataset is a public, token-accounted measurement of that routing decision, including the 0.3.0 in-memory cache and local retain rules.
This is not a production-log study. Labels come from Loghub. HDFS labels are block-level, then joined onto every line that… See the full description on the dataset page: https://huggingface.co/datasets/reachjalil/jevlogs-log-triage-benchmark.Open-Jev
Open-Jev: typed decision datasets
Open-Jev turns a state and a question into a typed decision: a yes/no probability, a distribution over choices, independent label probabilities, or a discrete numeric/ordinal decision. This repository publishes twelve separate, frozen data configs from the Open-Jev project, together with original manifests, exact raw records, source code and reconstruction instructions.
These are controlled, mostly synthetic tasks and reference labels. They are… See the full description on the dataset page: https://huggingface.co/datasets/ZefanCai/Open-Jev.open-jev-laya-benchjev-distill-corpus
Jev-Gate Student B — LoRA Memory-Relevance Judge
A LoRA adapter (r=16, α=32, on q_proj/v_proj) on Qwen/Qwen2.5-0.5B-Instruct, distilled from the Jev typed-judgment API into a compact local judge for agent-memory gating.
What it does: given a query and a candidate memory passage, outputs P(relevant) as the calibrated yes probability read from the final-token logits of yes vs no. Used to filter which vector-recalled memories get injected into agent context (vector recall →… See the full description on the dataset page: https://huggingface.co/datasets/SargeDev/jev-distill-corpus.INSTRUCT_JEV
INSTRUCT_JEV
INSTRUCT_JEV is an instruction corpus built from the TypeSafe AI documentation
for Jev, the first System One model. It is structured around the three TypeSafe
question primitives - Choice, Noul and Score - and mirrors the raw corpus
captured in deckerGUI-jev_corpus_RAW.
Credits
INSTRUCT_JEV is a DeckerGUI project and exists because of the work below.
Who
Contribution
Link
TypeSafe AI
Jev - the first System One model - and the Choice / Noul… See the full description on the dataset page: https://huggingface.co/datasets/ctaxnagomi/INSTRUCT_JEV.DGUI_HYPERMEM-JEV
DGUI_HYPERMEM-JEV
The training "brain" for DGUI-HyperMem (DeckerGUI HyperMemory) — the self-hosted
memory MCP server. Every JEV reasoning decision the service makes is appended here as a
typed instruction row, so the corpus grows with real usage and can be used to fine-tune or
few-shot the JEV layer later.
Usage
from datasets import load_dataset
ds = load_dataset("ctaxnagomi/DGUI_HYPERMEM-JEV", split="train")
for row in ds.stream():
print(row["use_case"]… See the full description on the dataset page: https://huggingface.co/datasets/ctaxnagomi/DGUI_HYPERMEM-JEV.poker-sft-mix-v1
HoldemMix-606k
HoldemMix-606k is a 606,630-record supervised-fine-tuning corpus for
heads-up no-limit Texas hold'em (HUNL 200BB). Each record is a chat-template
exchange in which an expert assistant decides the next action at a fully
specified poker decision node, optionally accompanied by an explicit
chain-of-thought rationale.
The corpus is the training data for
jevonmao/llama31-8b-poker-mix-v1-step10k,
known as PokerLlama-4, the 8B fine-tune that accompanies the
PokerLlama-4… See the full description on the dataset page: https://huggingface.co/datasets/jevonmao/poker-sft-mix-v1.jev-tree-choice-cap
Jev-tree vs the 255 choice cap
Synthetic multi-region incident catalog with 320 leaves. TypeSafe’s Jev can list 255 options in one choice. jev-tree walks an authored region→service→mode tree (or auto-partitions a flat list) so each call stays under the cap.
This is not production traffic and not the Loghub or PagerDuty studies.
Write-up: https://github.com/reachjalil/jev-tree/blob/main/docs/article/jev-tree-choice-cap.md
Code:… See the full description on the dataset page: https://huggingface.co/datasets/reachjalil/jev-tree-choice-cap.mini-jev-runs
mini-Jev run records: 27 900 schema-driven decisions with full candidate logits
Every record is one decision a frozen Qwen/Qwen3-4B-Instruct-2507 made about one field of a JSON
schema that arrived with the request. The field was turned into a lettered multiple-choice question
(A = pay_bill, B = bill_balance, …), the model ran one forward pass, and the answer was
read from its next-token logits over the option letters. No token was generated.
The records keep what such a run… See the full description on the dataset page: https://huggingface.co/datasets/Mikhail/mini-jev-runs.buildcanada-2025
BuildCanada 2025 Election Promises
Dataset Description
This repository contains campaign promises made by candidates in the 2025 Canadian Federal Election, formatted as markdown files.
Last Updated: 2026-06-14
Dataset Statistics
Total Candidates: 2
Total Promises: 1264
Total Promise Groups: 251
Dataset Structure
This repository is organized into the following directories:
candidates/ - Contains one file per candidate with all their… See the full description on the dataset page: https://huggingface.co/datasets/jevon/buildcanada-2025.laya-jev-benchmark
Laya vs Jev
TypeSafe released Jev on 15 September 2026, a closed Model that returns typed
Decisions instead of Text. Three Days later an open Reproduction appeared,
Laya (convaiinnovations/laya, Apache 2.0, 421M).
Laya's Model Card claims 83.8% against Jev's 67.8% and calls it a "+16.0%
Advantage". Those two Numbers are from two different Benchmarks, so the
Comparison says nothing.
I ran both on Benchmarks where Jev has published Numbers. One RTX 5090.
Everything below is… See the full description on the dataset page: https://huggingface.co/datasets/Luni/laya-jev-benchmark.tape-testaave_matchedje-veux-un-pack-de-computer-vision-pour-detecter-les-objets
Je veux un pack de computer vision pour detecter les objets,…
Je veux un pack de computer vision pour detecter les objets, dans des salles de bains pour le segment home avec des depth map
This dataset mirrors public data-pack render outputs from Physicl.
Each row represents one render view. The image column contains a stable URL to the primary render image uploaded under /data; image_path stores the relative repository path and data_commit_sha pins the Hugging Face dataset… See the full description on the dataset page: https://huggingface.co/datasets/physicl-test/je-veux-un-pack-de-computer-vision-pour-detecter-les-objets.JevForge-Mind2Web
JevForge Mind2Web Gold Decisions
Private research snapshot of gold candidate-decision records used by
JevForge.
What is included
Each JSONL record contains a page state, a choice question over candidate
elements, a noul question about one candidate, complete gold target
distributions, a website group, and source annotation metadata.
Split
Records
Websites
train
4,642
49
dev
786
6
calibration
400
6
test
800
8
ood
386
4
The 7,014 record IDs… See the full description on the dataset page: https://huggingface.co/datasets/AndeyTait/JevForge-Mind2Web.Jevanhielle_Zyhamont10_30postflop-solver-reasoning-traces-1m
Postflop-Solver Reasoning Traces (1M, v2)
Teacher-forced chain-of-thought reasoning traces for Heads-Up No-Limit Texas
Hold'em postflop decisions, distilled from a GTO solver (postflop-solver)
plus a strong LLM teacher.
Each example pairs a game scenario with the known-optimal solver action and
a step-by-step natural-language justification of why that action is correct.
The teacher is conditioned on the gold action (teacher forcing), so every trace
supports the correct move —… See the full description on the dataset page: https://huggingface.co/datasets/jevonmao/postflop-solver-reasoning-traces-1m.gtow-llama-sft-v3
GTO Wizard — Heads-Up NL Hold'em 200BB — SFT dataset (v3)
Supervised fine-tuning data for heads-up No-Limit Texas Hold'em, 200 big
blinds deep. Each row is a single decision point: a natural-language
description of the game state, paired with the game-theory-optimal action
GTO Wizard chose in that spot.
Intended for instruction-tuning a chat LLM to play HU 200BB poker (see the
pokerbench agent it was built for).
Schema
Two flat columns:
Column
Description… See the full description on the dataset page: https://huggingface.co/datasets/jevonmao/gtow-llama-sft-v3.jevlite_dataset
jevlite — synthetic decision questions with soft labels
5,866 typed questions about 978 synthetic program states — support tickets, SIEM alerts,
vendor invoices, AI agent transcripts, code reviews, incident logs, chat threads and job
applications — each answered by a teacher model as a full probability distribution
rather than a single label.
This is the synthetic portion of the training data for
vagmi/jev-lite, a decision model that reads a state
and a typed question and… See the full description on the dataset page: https://huggingface.co/datasets/vagmi/jevlite_dataset.Jevanhielle_Zyhamont_outCheckedjva-missions-report-raw
Dataset Card for "jva-missions-report-raw"
More Information needed
jev-rag-benchmark
Jev RAG Benchmark (English)
Frozen-candidate-pool evaluation of TypeSafe Jev 1.13 as the reranking
and decision layer of a RAG pipeline, compared with the free open-weights
OpenJev model and a NVIDIA cross-encoder, on English XQuAD and
SciFact. Every published run used free tiers (total cost: $0).
This repository contains the raw per-query artifacts, per-run reports with
paired bootstrap confidence intervals, calibration tables, and a plain-text
leaderboard.… See the full description on the dataset page: https://huggingface.co/datasets/emretheus/jev-rag-benchmark.finetuning_demojev-my-bro-dataset
jev-my-bro Governance Dataset
Provenance-aware English/Thai dataset for training and evaluating the typed
jev-my-bro decision model. Each case asks four structured governance
questions about an operation:
action: execute, ask_user, or reject
needs_review: whether explicit human review/approval is required
prohibited: whether the operation should be prohibited
risk: five-level operational risk
This snapshot contains 8,508 cases / 34,032 typed decisions.
This is… See the full description on the dataset page: https://huggingface.co/datasets/JonusNattapong/jev-my-bro-dataset.NoLimitHUPokerpokemon-with-pokedex-descriptions
Dataset Card for "pokemon-with-pokedex-descriptions"
More Information needed
tape-poc