SAVRN
Search Contact SAVRN

Open-weight model · Text classification

Jev-LCT-Qwen2.5-0.5B

by CaoHaoWei CaoHaoWei/Jev-LCT-Qwen2.5-0.5B

Jev-LCT-Qwen2.5-0.5B is an open-weight model for text classification from CaoHaoWei, released under Apache License 2.0. It has 494M parameters and a 32,768-token context. At 16-bit it needs about 1.2 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.

Jev-LCT-Qwen2.5-0.5B is the ultra-lightweight edge edition of the Jev-LCT System-One decision family. Weighing only ~1.9 GB in full bfloat16 weights, it is optimized for high-throughput API gateway routing, edge robotics, Raspberry Pi, and mobile deployment.

Parameters494M
Context32,768
Weights2.0 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads—

Runs On

What it takes to serve Jev-LCT-Qwen2.5-0.5B (494M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 1.0 GB 1.2 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 0.5 GB 0.6 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 0.2 GB 0.3 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 1, 2026.

Jev-LCT-Qwen2.5-0.5B on every accelerator the SAVRN Index prices, at every precision

Model Card

By CaoHaoWei, published under apache-2.0, revision 5cb5a7f032d4.

Jev-LCT-Qwen2.5-0.5B is the ultra-lightweight edge edition of the Jev-LCT System-One decision family. Weighing only ~1.9 GB in full bfloat16 weights, it is optimized for high-throughput API gateway routing, edge robotics, Raspberry Pi, and mobile deployment. - ~50ms 极低延迟：专为高并发 API 网关路由、实时内容审核设计； - MMLU 达 50.0%：大幅超越参数相近的判别模型（Laya 33.3%, Open-Jev 35.0%）； Apache License 2.0. Full repository at GitHub.

Read CaoHaoWei's full model card

Jev-LCT-Qwen2.5-0.5B: Edge Open System-One Decision Engine

"Decisions, Not Strings" meets "Free Calibrated Confidence"
《Jev-LCT-0.5B：端侧超轻量系统一决策引擎，专为高吞吐网关与边缘设备设计》

English Overview

Jev-LCT-Qwen2.5-0.5B is the ultra-lightweight edge edition of the Jev-LCT System-One decision family. Weighing only ~1.9 GB in full bfloat16 weights, it is optimized for high-throughput API gateway routing, edge robotics, Raspberry Pi, and mobile deployment.

Benchmark Highlights (300 Items on RTX 3090 Ti)

  • Intent Recognition (Banking77): 91.7%
  • Academic Multi-Task (MMLU): 50.0% (outperforming Laya 421M at 33.3% and Open-Jev at 35.0%)
  • Average Latency: 50.8 ms
  • Average Recurrent Loops: 1.36 loops

Quickstart

from lct_qwen_standalone import LCTQwen

model = LCTQwen.from_pretrained("CaoHaoWei/Jev-LCT-Qwen2.5-0.5B")

result = model.predict_choice(
    prompt="Classify customer query intent: 'I lost my debit card abroad'",
    choices=["card_lost", "wire_transfer", "balance_inquiry", "dispute_charge"]
)
print(f"Decision: {result['choice']} (Confidence: {result['confidence']:.2%})")

中文简介

  • 极致轻量：全重仅 1.9 GB，轻松常驻端侧与低显存边缘设备；
  • ~50ms 极低延迟：专为高并发 API 网关路由、实时内容审核设计；
  • MMLU 达 50.0%：大幅超越参数相近的判别模型（Laya 33.3%, Open-Jev 35.0%）；
  • 全量独立权重：开箱即用，支持单文件无依赖推理。

Citation & License

Apache License 2.0. Full repository at GitHub.

Configuration

Architecture
Qwen2ForCausalLM
Context length (tokens)
32,768
Layers
24
Hidden size
896
Feed-forward size
4,864
Attention heads
14
Key/value heads
2
Vocabulary size
151,936
RoPE base
1e+06
Model type
qwen2

Identity and Version

Repository
CaoHaoWei/Jev-LCT-Qwen2.5-0.5B
Publisher
CaoHaoWei
Task
Text classification
Modality
Text
Library
transformers
Parameters
494M parameters
Languages
en, zh
Revision
5cb5a7f032d4459d32196060895e441aef285139
First published
2026-09-25
Last updated
2026-09-26

Files and Weights

14 files, 2.0 GB in total. The weights are 2 files totalling 2.0 GB in pt, safetensors.

Weights2 files · 2.0 GB
Configuration5 files · 3.1 KB
Tokenizer4 files · 15.9 MB
Documentation1 file · 3.0 KB
Other1 file · 2.5 KB
Repository1 file · 1.6 KB
Every file
FileTypeSizeSHA-256
lct_auxiliary.ptWeights24.0 KB d54936a1d3bc
model.safetensorsWeights2.0 GB 42d54e30942d
added_tokens.jsonConfiguration629 B —
config.jsonConfiguration1.3 KB —
generation_config.jsonConfiguration123 B —
lct_config.jsonConfiguration398 B —
special_tokens_map.jsonConfiguration647 B —
README.mdDocumentation3.0 KB —
chat_template.jinjaOther2.5 KB —
.gitattributesRepository1.6 KB —
merges.txtTokenizer1.7 MB —
tokenizer.jsonTokenizer11.4 MB 9c5ae00e602b
tokenizer_config.jsonTokenizer4.9 KB —
vocab.jsonTokenizer2.8 MB —

License and Download

License
apache-2.0
Access
Open weights, no gate
Download size
2.0 GB
Download from CaoHaoWei

Released by CaoHaoWei through its official repository on Hugging Face. Read the license.

Built From

  • Trained on (disclosed) ai2_arc
  • Trained on (disclosed) banking77
  • Trained on (disclosed) boolq
  • Trained on (disclosed) mmlu
  • Trained on (disclosed) truthful_qa

Memory Requirements

PrecisionWeights in memory
As published2.0 GB
16-bit1.0 GB
8-bit0.5 GB
4-bit0.2 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About Jev-LCT-Qwen2.5-0.5B

How much GPU memory does Jev-LCT-Qwen2.5-0.5B need?

About 1.2 GB at 16-bit and 0.3 GB at 4-bit: the weights (494M parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run Jev-LCT-Qwen2.5-0.5B on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use Jev-LCT-Qwen2.5-0.5B commercially?

Yes. Jev-LCT-Qwen2.5-0.5B is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

What is Jev-LCT-Qwen2.5-0.5B's context length?

32,768 tokens, from the maximum position embeddings in its published configuration.

Similar Models

Opir-multitask-large is the English, highest-accuracy multi-task checkpoint in the Opir family: an encoder-based GLiClass guardrail model for real-time LLM safety filtering. It supports binary safe/unsafe classification, toxicity detection, jailbreak and prompt-injection detection, and zero-shot harmful-content categorization over a hierarchical safety taxonomy. This card is for knowledgator/opir-multitask-large. The model is used through GLiClass zero-shot classification: pass text plus the candidate labels you want scored. Use single-label mode for binary safe/unsafe decisions and multi-label mode for taxonomy, toxicity, jailbreak, or custom policy labels. Use multi-label mode when you…

Open weights apache-2.0 439M parameters gliclass
View model

Model · Text classification

erabi-practical-v1-experimental

Sugarknight

This is an experimental, uncalibrated choice-ranking model. It is not an official Jev model, a validated general-purpose reasoner, or an automatic decision-maker. The model ranks 2–16 user-supplied candidate texts for a natural-language context and question and returns all candidate probabilities through the ERABI code. Decisions should be reviewed by a person. - Practical V1 data consists of original synthetic Japanese, English, and Simplified Chinese examples in six task families, generated and answer-blind rejudged with DeepSeek V4.1 Flash. - The Exam-QA source was filtered and transformed with the same DeepSeek model. Symbolic answer labels were mapped to source choice text. Ambiguous…

Open weights apache-2.0 439M parameters transformers
View model

Model · Text classification

deberta_MP_dynamic

Oriane Peter

This model is a fine-tuned version of microsoft/deberta-v3-large on an unknown dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 3e-05 - trainbatchsize: 16 - evalbatchsize: 16 - gradientaccumulationsteps: 2 - totaltrainbatchsize: 32 - lrschedulertype: cosine - lrschedulerwarmupsteps: 0.1 - numepochs: 1000 - Transformers 5.12.1 - Pytorch 2.11.0+cu128 - Datasets 5.0.1 - Tokenizers 0.22.2

Open weights mit 435M parameters 512 tokens transformers
View model

Model · Text classification

DEBATE-kor-large

Jong Rock Jeong

DEBATE-kor-large is a Korean-adapted Political DEBATE model for binary natural language inference (NLI) on political text. The model is initialized from mlburnham/PoliticalDEBATEDeBERTalargev1.1, the original DeBERTa-based Political DEBATE checkpoint, and subsequently fine-tuned on jongrock17/PolNLI-kor, a Korean translation and adaptation of PolNLI. Political DEBATE DeBERTa-large → PolNLI-kor → DEBATE-kor-large Unlike the PolNLI-kor-RoBERTa model family, which starts from Korean-pretrained KLUE-RoBERTa encoders, DEBATE-kor directly adapts the original Political DEBATE checkpoint to Korean political NLI. DEBATE-kor formulates NLI as a binary classification problem. notentailment combines…

Open weights 435M parameters 512 tokens transformers
View model

Model · Text classification

laya-typed-decisions

Convai Innovations

Non-autoregressive System 1 decision model, fine-tuned on the typed-decisions workflows: agent-trace observability, customer service, invoice processing and security incidents. Part of the Laya family. 400 test cases, 2,000 decisions, measured on the official test split. +3.9 points over Jev's published 0.727, above the 0.735 teacher ceiling, with 2.4x better Brier and 1.6x better score MAE. Jev figures are third-party published, not measured here — there is no TypeSafe API access in this project, and sample sizes and prompts differ. Treat the comparison as indicative. Router will not select this checkpoint automatically unless you construct it with autotaskdetection=True — it is…

Open weights apache-2.0 421M parameters transformers
View model

Model · Text classification

laya

Convai Innovations

Multilingual, non-autoregressive System 1 decision model. Give it a state (text, email, ticket, or JSON) and typed questions; it returns typed answers with mathematically calibrated probabilities in a single forward pass (~33 ms) across 100+ languages. Trained with reinforcement learning against strictly proper scoring rules (RLCD), so reporting honest probabilities is the only way to maximise reward. It never generates text, so there is nothing to parse and nothing to hallucinate. This repo holds all three checkpoints and is the hub for the family. The English checkpoint is at the repo root; the other two are bundled subfolders, and only the one you request is downloaded: pip install -U…

Open weights apache-2.0 421M parameters transformers
View model