How much memory does LFM2.5-VL-1.6B need?

About 2.5 GB at 4-bit, 3.3 GB at 8-bit and 4.8 GB at full 16-bit precision, including an 8,192-token KV cache and about 1.5 GB of runtime overhead. These are estimates from its parameter count.

Can LFM2.5-VL-1.6B run on a 24 GB graphics card?

Yes, at BF16 by this estimate, with room left for context.

Is LFM2.5-VL-1.6B free for commercial use?

Its card lists a custom licence (lfm1.0), which sets its own terms; read it before any commercial use. Read the licence itself before relying on this summary; it is not legal advice.

What is LFM2.5-VL-1.6B’s context window?

Its configuration sets 128,000 tokens. Longer contexts need more memory for the KV cache, so the practical limit on a given machine can be lower.

ANVE: the private AI office, on the record. See what that means

ANVE

Run it on your own hardware · updated 2026-10-16

LFM2.5-VL-1.6B
Memory at each precision, the licence in plain words, and where it fits.

Short answer. LFM2.5-VL-1.6B needs about 2.5 GB of memory at 4-bit, 3.3 GB at 8-bit and 4.8 GB at full 16-bit precision, including runtime overhead and an 8,192-token KV cache. It has about 1.6 billion parameters and a 128,000-token context window. Licence in short: custom licence: check before commercial use. By memory alone, on ANVE’s launch configurations: ANVE One fits at BF16, ANVE Office fits at BF16, ANVE Office Pro (GPU) fits at BF16. These are estimates from its published size, not benchmarks.

What it is

LFM2.5-VL-1.6B is a vision-language model: it reads images (scanned pages, photos, charts and screenshots) as well as text, published by Liquid AI on Hugging Face on 5 January 2026. It has about 1.6 billion parameters and a configured context window of 128,000 tokens.

Its card lists the licence as a custom licence (lfm1.0). In short, the licence sets its own terms; read it before any commercial use.

Hugging Face counted about 30,609 downloads of it in the 30 days before we checked (7 Oct 2026), and 330 people have liked it there.

The card lists 8 languages: en, ja, ko, fr, es, de, ar, zh (ISO codes as given on the card).

It is built on LiquidAI/LFM2.5-1.2B-Base, according to its card.

How much memory it needs

The arithmetic: parameters × bits per weight ÷ 8 gives the weights; add the KV cache, which grows with every token of context; then add about 1.5 GB for the runtime. These are estimates for planning, not measurements.

Each token of context adds about 0.01 MB of KV cache at 16-bit, so 32,768 tokens would add about 0.40 GB. Only 6 of its 16 layers use standard attention; the others keep a small fixed state, so the cache is smaller than in a classic transformer.

Precision	Weights	KV cache (8,192 tokens)	Total, estimated
BF16 (full precision)	3.2 GB	0.10 GB	4.8 GB
8-bit (Q8_0)	1.7 GB	0.10 GB	3.3 GB
4-bit (Q4_K)	0.90 GB	0.10 GB	2.5 GB

Architecture at a glance

Setting	Value in config.json
Architecture class	Lfm2VlForConditionalGeneration
Model type	lfm2_vl
Layers	16
Hidden size	2,048
Attention heads	32
Key-value heads	8 (grouped-query attention)
Head size	64
Vocabulary	65,536
Configured context	128,000
Stored precision	bfloat16

Will it fit on ANVE?

How the verdicts are worked out: on ANVE One, about three quarters of 64 GB of unified memory is available to the model; on ANVE Office and ANVE Office Pro, the model has to fit in the 24 GB on the graphics card, with 64 GB or 128 GB of system memory for everything else. This is arithmetic against the launch specification, not a test. ANVE’s tested default is Qwen 3.8 27B; this model is not yet validated on ANVE.

Configuration (at launch)	Memory for the model	Verdict
ANVE One	about 48 GB of 64 GB unified memory	fits at BF16
ANVE Office	24 GB on the RTX PRO 4000 Blackwell SFF Edition	fits at BF16
ANVE Office Pro (GPU)	24 GB on the RTX PRO 4000 Blackwell	fits at BF16

What an office would use it for

Models this small are best for routing, classification, autocomplete and simple extraction. They are not reliable for minutes or legal drafting: ANVE never uses anything smaller than a 4B-class model for minutes.

Similar models to compare

MiniCPM-V-4.6 (1.3B parameters, about 2.2 GB at 4-bit before context; commercial use allowed)
Qwen3-VL-2B-Instruct (2.1B parameters, about 2.7 GB at 4-bit before context; commercial use allowed)
InternVL2-2B (2.2B parameters, about 2.7 GB at 4-bit before context; commercial use allowed)

Questions people ask

How much memory does LFM2.5-VL-1.6B need?: About 2.5 GB at 4-bit, 3.3 GB at 8-bit and 4.8 GB at full 16-bit precision, including an 8,192-token KV cache and about 1.5 GB of runtime overhead. These are estimates from its parameter count.
Can LFM2.5-VL-1.6B run on a 24 GB graphics card?: Yes, at BF16 by this estimate, with room left for context.
Is LFM2.5-VL-1.6B free for commercial use?: Its card lists a custom licence (lfm1.0), which sets its own terms; read it before any commercial use. Read the licence itself before relying on this summary; it is not legal advice.
What is LFM2.5-VL-1.6B’s context window?: Its configuration sets 128,000 tokens. Longer contexts need more memory for the KV cache, so the practical limit on a given machine can be lower.

What is not real yet

No sealed ANVE unit exists; every image of the appliance on this site is a labelled render. The software runs today on a development machine. No customers yet, no certifications, no delivery date, and the final price is not announced.

Reservations are ₹999 in India or US$12 elsewhere, refundable in full until you confirm an order in writing (terms). Offices that cannot use cloud AI can apply to be one of ten design partners.

LFM2.5-VL-1.6BMemory at each precision, the licence in plain words, and where it fits.