What are the hardware requirements for Catlm?

Catlm requires approximately 7.8 GB of VRAM and supports a context window of 8K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Catlm and how large is it?

Catlm is developed by nikitharao. 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 Catlm?

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

Catlm by nikitharao — VRAM 7.8GB, 8K context

Name: Catlm
Author: nikitharao

Catlm is an open-source language model by nikitharao. Features: LLM, VRAM: 7.8GB, Context: 8K, License: mit, LLM Explorer Score: 0.09.

Arxiv:2310.01602 Endpoints compatible Gpt neox Pytorch Region:us

Model Card on HF 🤗: https://huggingface.co/nikitharao/catlm

Catlm Parameters and Internals

Model Type

GPT-style, Language Model

Additional Notes

The model was trained without the ~~ token and should be removed.

Training Details

Data Sources:

Python projects, Java projects

Data Volume:

~260GB

Methodology:

Novel pretraining signal that considers the mapping between code and test files

Context Length:

8192

Model Architecture:

GPT-style

LLM Name	Catlm
Repository 🤗	https://huggingface.co/nikitharao/catlm
Required VRAM	7.8 GB
Updated	2026-08-08
Maintainer	nikitharao
Model Type	gpt_neox
Model Files	7.8 GB
Model Architecture	GPTNeoXForCausalLM
License	mit
Context Length	8192
Model Max Length	8192
Transformers Version	4.24.0
Tokenizer Class	T5Tokenizer
Padding Token	<pad>
Vocabulary Size	64000
Torch Data Type	float16

Rank the Catlm 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 Catlm

Best Alternatives	Context / RAM	Downloads	Likes
...Prover 14final Checkpoint 5830	4K / 14.9 GB	6	0
Neox Musenet Untrained	4K / 7.3 GB	6	0
Stabillm Instruct De	4K / 31.8 GB	7	0
MusePy 1 1	2K / 0.7 GB	25	1
Open Calm Large	2K / 1.8 GB	1346	10
MonoCoder OMP	2K / 3.6 GB	13	0
ProofGPT V0.1	2K / 2.9 GB	76	3
KULLM RLHF	2K / 25.8 GB	11	3
Open Calm Small	2K / 0.4 GB	2214	20
Step3 Mk7	2K / 25.8 GB	7	0

Note: green Score (e.g. "73.2") means that the model is better than nikitharao/catlm.