Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · NECOUDBFM

Jellyfish-8B

Jellyfish-8B is an 8.03B-parameter Llama text-generation model from NECOUDBFM. According to the model card, it was fine-tuned from Meta-Llama-3-8B-Instruct for data preprocessing work.

Publisher
NECOUDBFM
Task
text-generation
Model type
llama
License
cc-by-nc-4.0
Library
transformers
Publication status
Accepted · not indexed

Model overview

Jellyfish-8B is published by NECOUDBFM as a Llama text-generation model. The captured configuration identifies LlamaForCausalLM with model type llama, and Safetensors metadata reports 8,030,261,248 parameters. According to the model card, it was fine-tuned from Meta-Llama-3-8B-Instruct on a subset of the Jellyfish-Instruct dataset and is described in the Jellyfish paper on data preprocessing.

Recorded capabilities

Data-preprocessing fine-tune

According to the model card, the model was fine-tuned from Meta-Llama-3-8B-Instruct on a Jellyfish-Instruct subset for preprocessing tasks.

LoRA training record

According to the model card, training used LoRA targeting the q_proj, k_proj, v_proj, and o_proj modules.

Paper-backed evaluation tables

According to the model card, seen and unseen task tables compare Jellyfish variants with prior methods; the card cites the Jellyfish paper arXiv 2312.01678.

Documented inference path

According to the model card, the publisher recommends vLLM for inference and provides Python and prompt-template examples.

Use cases in the source record

  • Data-preprocessing tasks such as error detection, data imputation, schema matching, and entity matching documented in the model card tables.
  • Column-type annotation and attribute-value extraction using the card's documented task prompts.
  • vLLM-accelerated inference following the publisher's documented vLLM and Python-script examples.

Limitations and unknowns

  • Card performance tables are publisher-reported comparisons without independently extracted benchmark verification in this record.
  • No context-window value, VRAM requirement, or quantization detail was extracted.
  • Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: NECOUDBFM/Jellyfish-8B

Captured: Unknown. Processed: 2026-09-07T19:34:34.612652+00:00.

Jellyfish-8B Jellyfish models with other sizes are available here: Jellyfish-7B Jellyfish-13B Model Details Jellyfish-8B is a large language model equipped with 8 billion parameters. We fine-tuned the Meta-Llama-3-8B-Instruct model using a subset of the Jellyfish-Instruct dataset. More details about the model can be found in the Jellyfish paper . Developed by: Haochen Zhang, Yuyang Dong, Chuan Xiao, Masafumi Oyamada Contact: dongyuyang@nec.com Funded by: NEC Corporation, Osaka University Language(s) (NLP): English License: Non-Commercial Creative Commons license (CC BY-NC-4.0) Finetuned from model: Meta-Llama-3-8B-Instruct Citation…

F001F002F003F004F005F006F007F008F009F010F012F013F014F015F016F017F020F021F022F025F028F031F032