What are the hardware requirements for Codellama 7B Hf ReFT GSM8k?

Codellama 7B Hf ReFT GSM8k requires approximately 13.5 GB of VRAM and supports a context window of 16K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Codellama 7B Hf ReFT GSM8k and how large is it?

Codellama 7B Hf ReFT GSM8k is developed by lqtrung1998, a model with 7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Codellama 7B Hf ReFT GSM8k perform on standard benchmarks?

Codellama 7B Hf ReFT GSM8k has the following published scores: MMLU 40.86. Compare against reference models on this page or on the LLM Explorer leaderboards.

Codellama 7B Hf ReFT GSM8k by lqtrung1998 — VRAM 13.5GB, 16K context

Name: Codellama 7B Hf ReFT GSM8k
Author: lqtrung1998

Codellama 7B Hf ReFT GSM8k is an open-source language model by lqtrung1998. Features: 7b LLM, VRAM: 13.5GB, Context: 16K, License: llama2, Code Generating, LLM Explorer Score: 0.14, Arc: 43.5, HellaSwag: 64.5, MMLU: 40.9, GSM8K: 23.7.

Arxiv:2401.08967 Codegen Endpoints compatible Llama Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/lqtrung1998/Codellama-7b-hf-ReFT-GSM8k

Codellama 7B Hf ReFT GSM8k Parameters and Internals

Model Type

Reasoning, Code Understanding

Use Cases

Areas:

Commercial, Research

Applications:

Code synthesis, Understanding tasks

Primary Use Cases:

Python programming language

Limitations:

Use is limited to English, Use that violates applicable laws, Outside scenarios covered in testing

Considerations:

Developers should perform safety testing and tuning.

Additional Notes

The models are tuned on Codellama and applicable licenses are inherited.

Supported Languages

English (Proficient)

Training Details

Data Sources:

GSM8k dataset

Methodology:

Reinforced Fine-Tuning after Supervised Fine-Tuning

Responsible Ai Considerations

Mitigation Strategies:

Developers should perform safety testing and tuning tailored to their specific applications of the model.

Input Output

Input Format:

Question followed by 'Answer reasoning:'

Accepted Modalities:

text

Output Format:

Code reasoning in Python format

LLM Name	Codellama 7B Hf ReFT GSM8k
Repository 🤗	https://huggingface.co/lqtrung1998/Codellama-7b-hf-ReFT-GSM8k
Model Size	7b
Required VRAM	13.5 GB
Updated	2026-06-25
Maintainer	lqtrung1998
Model Type	llama
Model Files	10.0 GB: 1-of-2 3.5 GB: 2-of-2
Generates Code	Yes
Model Architecture	LlamaForCausalLM
License	llama2
Context Length	16384
Model Max Length	16384
Transformers Version	4.34.1
Tokenizer Class	CodeLlamaTokenizer
Padding Token	<s>
Vocabulary Size	32016
Torch Data Type	bfloat16

Rank the Codellama 7B Hf ReFT GSM8k 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 Codellama 7B Hf ReFT GSM8k

Best Alternatives	Context / RAM	Downloads	Likes
Deepseek Coder 6.7B Instruct	16K / 13.5 GB	535311	507
CodeLlama 7B Hf	16K / 13.5 GB	254150	377
CodeLlama 7B Instruct Hf	16K / 13.5 GB	58587	258
GetCode Slerp	16K / 13.6 GB	87	1
Wizardllama 7B	16K / 13.5 GB	48	0
GoLLIE 7B	16K / 40.5 GB	2779	32
MathCoder2 CodeLlama 7B	16K / 13.5 GB	10	5
Stack Codellama 7B Inst	16K / 13.5 GB	66	0
CodeLlama 7B Python Hf	16K / 13.5 GB	1935	146
CodeV CL 7B	16K / 13.5 GB	430	1

Note: green Score (e.g. "73.2") means that the model is better than lqtrung1998/Codellama-7b-hf-ReFT-GSM8k.