What are the hardware requirements for LFM2.5 24B A2B NVFP4A16?

LFM2.5 24B A2B NVFP4A16 requires approximately 0.3 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed LFM2.5 24B A2B NVFP4A16 and how large is it?

LFM2.5 24B A2B NVFP4A16 is developed by primitive-ai, a model with 24b 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 LFM2.5 24B A2B NVFP4A16?

LFM2.5 24B A2B NVFP4A16 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

LFM2.5 24B A2B NVFP4A16 by primitive-ai — VRAM 0.3GB, 128K context

Name: LFM2.5 24B A2B NVFP4A16
Author: primitive-ai

LFM2.5 24B A2B NVFP4A16 is an open-source language model by primitive-ai. Features: 24b LLM, VRAM: 0.3GB, Context: 128K, License: other, LLM Explorer Score: 0.31.

8-bit Agentic Ar Base model:liquidai/lfm2.5-24b... Base model:quantized:liquidai/... Compressed-tensors Conversational De En Es Fr Hi Id It Ja Ko Lfm2.5 Lfm2 moe Liquid Moe Nvfp4 Pl Pt Quantized Reasoning Region:us Ru Safetensors Single-gpu Th Tool-calling Vi Vllm Zh

Model Card on HF 🤗: https://huggingface.co/primitive-ai/LFM2.5-24B-A2B-NVFP4A16

LFM2.5 24B A2B NVFP4A16 Parameters and Internals

LLM Name	LFM2.5 24B A2B NVFP4A16
Repository 🤗	https://huggingface.co/primitive-ai/LFM2.5-24B-A2B-NVFP4A16
Base Model(s)	LiquidAI/LFM2.5-24B-A2B LiquidAI/LFM2.5-24B-A2B
Model Size	24b
Required VRAM	0.3 GB
Updated	2026-10-20
Maintainer	primitive-ai
Model Type	lfm2_moe
Model Files	2.0 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB 0.3 GB
Supported Languages	en ar zh fr de ja ko es pt it vi th id hi ru pl
Model Architecture	Lfm2MoeForCausalLM
License	other
Context Length	131072
Model Max Length	131072
Transformers Version	5.12.1
Tokenizer Class	TokenizersBackend
Padding Token	<\|pad\|>
Vocabulary Size	128000

Rank the LFM2.5 24B A2B NVFP4A16 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 LFM2.5 24B A2B NVFP4A16

Best Alternatives	Context / RAM	Downloads	Likes
LFM2.5 24B A2B	128K / 48.3 GB	9356	424
LFM2.5 24B A2B FP8	128K / 24.6 GB	1691	22
LFM2.5 24B A2B Base	128K / 48.3 GB	362	35
... 24B A2B DERISKED EXL3 3.25bpw	128K / 10.4 GB	130	1
...demLLM LFM2.5 24B A2B EXL3 2.5	128K / 8.2 GB	84	1
...demLLM LFM2.5 24B A2B EXL3 3.0	128K / 9.7 GB	82	1
LFM2 24B A2B	125K / 47.7 GB	13081	340
LFM2 24B A2B ONNX	125K / GB	98	13
LFM2 24B A2B MLX Bf16	125K / 48 GB	80	7
LFM2.5 24B A2B MLX 6bit	128K / 19.5 GB	828	2

Note: green Score (e.g. "73.2") means that the model is better than primitive-ai/LFM2.5-24B-A2B-NVFP4A16.