What are the hardware requirements for Speck2 140M Instruct?

Speck2 140M Instruct requires approximately 0.3 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Speck2 140M Instruct and how large is it?

Speck2 140M Instruct is developed by specklabs, a model with 140m 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 Speck2 140M Instruct?

Speck2 140M Instruct is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Speck2 140M Instruct by specklabs — VRAM 0.3GB, 4K context

Name: Speck2 140M Instruct
Author: specklabs

Speck2 140M Instruct is an open-source language model by specklabs. Features: 140m LLM, VRAM: 0.3GB, Context: 4K, License: mit, Instruction-Based, LLM Explorer Score: 0.27.

Base model:finetune:specklabs/... Base model:specklabs/speck2-14... Causal-convolution Chat Conversational Custom code Dataset:huggingfacetb/finemath Dataset:huggingfacetb/smollm-c... Dataset:mlfoundations/dclm-bas... Dataset:openbmb/ultra-fineweb Dataset:openbmb/ultra-fineweb-... Dataset:specklabs/speckchat2 Dataset:wikimedia/wikipedia En Grouped-query-attention Hybrid Instruct Instruction-tuned Pytorch Region:us Safetensors Speck

Model Card on HF 🤗: https://huggingface.co/specklabs/Speck2-140M-Instruct

Speck2 140M Instruct Parameters and Internals

LLM Name	Speck2 140M Instruct
Repository 🤗	https://huggingface.co/specklabs/Speck2-140M-Instruct
Base Model(s)	Speck2 140M specklabs/Speck2-140M
Model Size	140m
Required VRAM	0.3 GB
Updated	2026-09-02
Maintainer	specklabs
Model Type	speck
Instruction-Based	Yes
Model Files	0.3 GB
Supported Languages	en
Model Architecture	SpeckForCausalLM
License	mit
Context Length	4096
Model Max Length	4096
Transformers Version	5.1.0
Tokenizer Class	SpeckTokenizer
Vocabulary Size	32003

Rank the Speck2 140M 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 Speck2 140M Instruct

Best Alternatives	Context / RAM	Downloads	Likes
Speck1 140M Instruct	4K / 0.3 GB	972	2
Speck1.1 140M Instruct	4K / 0.3 GB	500	3
Speck1.5 140M Instruct	4K / 0.3 GB	123	1

Note: green Score (e.g. "73.2") means that the model is better than specklabs/Speck2-140M-Instruct.