What are the hardware requirements for Starcoder Gpteacher Code Instruct?

Starcoder Gpteacher Code Instruct requires approximately 63.3 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Starcoder Gpteacher Code Instruct and how large is it?

Starcoder Gpteacher Code Instruct is developed by GeorgiaTechResearchInstitute. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Starcoder Gpteacher Code Instruct perform on standard benchmarks?

Starcoder Gpteacher Code Instruct has the following published scores: MMLU 28.63. Compare against reference models on this page or on the LLM Explorer leaderboards.

Starcoder Gpteacher Code Instruct by GeorgiaTechResearchInstitute — VRAM 63.3GB

Name: Starcoder Gpteacher Code Instruct
Author: GeorgiaTechResearchInstitute

Starcoder Gpteacher Code Instruct is an open-source language model by GeorgiaTechResearchInstitute. Features: LLM, VRAM: 63.3GB, License: bigcode-openrail-m, Instruction-Based, Code Generating, LLM Explorer Score: 0.11, Arc: 32.7, HellaSwag: 47.6, MMLU: 28.6.

Arxiv:1911.02150 Arxiv:2205.14135 Arxiv:2207.14255 Arxiv:2305.06161 Code-gen Codegen Dataset:bigcode/the-stack-dedu... Dataset:teknium1/gpteacher-cod... Endpoints compatible Gpt bigcode Instruct Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/GeorgiaTechResearchInstitute/starcoder-gpteacher-code-instruct

Starcoder Gpteacher Code Instruct Parameters and Internals

Model Type

text-generation

Use Cases

Primary Use Cases:

Code generation

Limitations:

The generated code might be inefficient, contain bugs or exploits. Current fine-tuning results may not represent the model's full potential.

Supported Languages

languages (80+ Programming languages), proficiency ()

Training Details

Data Sources:

bigcode/the-stack-dedup, teknium1/GPTeacher-codegen

Data Volume:

1 trillion tokens

Methodology:

Multi Query Attention, Fill-in-the-Middle objective

Context Length:

8192

Model Architecture:

GPT-2 model with multi-query attention and Fill-in-the-Middle objective

LLM Name	Starcoder Gpteacher Code Instruct
Repository 🤗	https://huggingface.co/GeorgiaTechResearchInstitute/starcoder-gpteacher-code-instruct
Required VRAM	63.3 GB
Updated	2026-08-08
Maintainer	GeorgiaTechResearchInstitute
Model Type	gpt_bigcode
Instruction-Based	Yes
Model Files	9.9 GB: 1-of-7 9.9 GB: 2-of-7 9.8 GB: 3-of-7 9.9 GB: 4-of-7 9.8 GB: 5-of-7 9.9 GB: 6-of-7 4.1 GB: 7-of-7 0.0 GB
Generates Code	Yes
Model Architecture	GPTBigCodeForCausalLM
License	bigcode-openrail-m
Model Max Length	1024
Transformers Version	4.28.1
Tokenizer Class	GPT2Tokenizer
Beginning of Sentence Token	<\|endoftext\|>
End of Sentence Token	<\|endoftext\|>
Unk Token	<\|endoftext\|>
Vocabulary Size	49153
Torch Data Type	float32
Activation Function	gelu
Errors	replace

Rank the Starcoder Gpteacher Code Instruct 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 Starcoder Gpteacher Code Instruct

Best Alternatives	Context / RAM	Downloads	Likes
...der Skeleton Wizard Coder 8bit	0K / 15.9 GB	3	1