What are the hardware requirements for CHEETAH 350M Merged FP16?

CHEETAH 350M Merged FP16 requires approximately 0.7 GB of VRAM and supports a context window of 125K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed CHEETAH 350M Merged FP16 and how large is it?

CHEETAH 350M Merged FP16 is developed by Tralalabs, a model with 350m 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 CHEETAH 350M Merged FP16?

CHEETAH 350M Merged FP16 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

CHEETAH 350M Merged FP16 by Tralalabs — VRAM 0.7GB, 125K context

Name: CHEETAH 350M Merged FP16
Author: Tralalabs

CHEETAH 350M Merged FP16 is an open-source language model by Tralalabs. Features: 350m LLM, VRAM: 0.7GB, Context: 125K, License: other, Quantized, Instruction-Based, Merged, LLM Explorer Score: 0.24.

Merged Model Base model:finetune:liquidai/l... Base model:liquidai/lfm2-350m Cheetah Cheetah-350m Conversational Dataset:huggingfacetb/smol-smo... En Endpoints compatible Fp16 Instruct Instruction-tuned Lfm2 Liquid-ai Modal Quantized Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/Tralalabs/CHEETAH-350M-Merged-FP16

CHEETAH 350M Merged FP16 Parameters and Internals

LLM Name	CHEETAH 350M Merged FP16
Repository 🤗	https://huggingface.co/Tralalabs/CHEETAH-350M-Merged-FP16
Base Model(s)	LFM2 350M LiquidAI/LFM2-350M
Merged Model	Yes
Model Size	350m
Required VRAM	0.7 GB
Updated	2026-07-10
Maintainer	Tralalabs
Model Type	lfm2
Instruction-Based	Yes
Model Files	0.7 GB
Supported Languages	en
Quantization Type	fp16
Model Architecture	Lfm2ForCausalLM
License	other
Context Length	128000
Model Max Length	128000
Transformers Version	5.9.0
Tokenizer Class	TokenizersBackend
Padding Token	<\|pad\|>
Vocabulary Size	65536

Rank the CHEETAH 350M Merged FP16 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 CHEETAH 350M Merged FP16

Best Alternatives	Context / RAM	Downloads	Likes
...ro Roleplay Instruct 350M 0812	125K / 0.2 GB	31	1
Palmer 005 Core	125K / 0.7 GB	5	1

Note: green Score (e.g. "73.2") means that the model is better than Tralalabs/CHEETAH-350M-Merged-FP16.