What are the hardware requirements for Results?

Results requires approximately 0.9 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Results and how large is it?

Results is developed by Noodle-bg, a model with 3b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Results?

Results is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Results by Noodle-bg — VRAM 0.9GB

Name: Results
Author: Noodle-bg

Results is an open-source language model by Noodle-bg. Features: 3b LLM, VRAM: 0.9GB, License: apache-2.0, LLM Explorer Score: 0.13.

Adapter Base model:adapter:openlm-rese... Base model:openlm-research/ope... Distilbert Finetuned Generated from trainer Lora Peft Region:us Safetensors Sft Tensorboard Trl

Model Card on HF 🤗: https://huggingface.co/Noodle-bg/results

Results Parameters and Internals

Additional Notes

The model is a fine-tuned version of the base model open_llama_3b_v2. Training procedure specifics include using a learning rate of 5e-05 with the Adam optimizer, and a cosine learning rate scheduler. Batch sizes for training and evaluation were set to 2 and 8, respectively, with total training conducted over 3 epochs.

LLM Name	Results
Repository 🤗	https://huggingface.co/Noodle-bg/results
Base Model(s)	Open Llama 3b V2 openlm-research/open_llama_3b_v2
Model Size	3b
Required VRAM	0.9 GB
Updated	2025-02-17
Maintainer	Noodle-bg
Model Files	0.9 GB 0.0 GB
Model Architecture	Adapter
License	apache-2.0
Model Max Length	2048
Is Biased	none
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
PEFT Type	LORA
LoRA Model	Yes
PEFT Target Modules	o_proj\|k_proj\|q_proj\|v_proj
LoRA Alpha	32
LoRA Dropout	0.05
R Param	16

Rank the Results capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Results

Best Alternatives	Context / RAM	Downloads	Likes
...Mental Support Qwen2.5 3B Lora	0K / 0.1 GB	21	0
Qwen25 3B Korean Pii Qlora2	0K / 0.1 GB	12	0
Llama3 Code Lora	0K / 0.1 GB	13	2
...ma 3.2 3B Legal India Qlora V2	0K / 0.1 GB	84	0
Nomi 1.1	0K / 6.5 GB	6	3
Art Skynet 3B	0K / 6.5 GB	11	15
...ull Lr5e4 Peft Mlp 32 32 Bs256	0K / 0.7 GB	5	0
Xenith 3B	0K / 7.6 GB	2	2
Accel2	0K / 6.8 GB	6	0
Accel3	0K / 0 GB	5	0

Note: green Score (e.g. "73.2") means that the model is better than Noodle-bg/results.