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

ANVE

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

nllb-200-distilled-600M
How much memory it needs, what its licence allows, and whether it fits an office box.

Short answer. nllb-200-distilled-600M needs about 1.8 GB of memory at 4-bit, 2.1 GB at 8-bit and 2.7 GB at full 16-bit precision, including runtime overhead. It has about 600 million parameters and a 1,024-token context window. Licence in short: non-commercial only. 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.

01

What it is

nllb-200-distilled-600M is a machine-translation model, published by Meta (FAIR) on Hugging Face on 8 July 2022. It has about 600 million parameters and a configured context window of 1,024 tokens.

Its card lists the licence as CC BY-NC 4.0. In short, the licence allows non-commercial use only, so an office cannot use it for paid client work without a separate licence.

Hugging Face counted about 819,618 downloads of it in the 30 days before we checked (7 Oct 2026), and 983 people have liked it there.

The card lists 196 languages: ace, acm, acq, aeb, af, ajp, ak, als, am, apc, ar, ars and 184 more (ISO codes as given on the card).

02

How much memory it needs

The arithmetic: parameters × bits per weight ÷ 8 gives the weights, plus about 1.5 GB for the runtime. These are planning estimates, not measurements.

Inputs are short, so working memory is small next to the weights.

PrecisionWeightsTotal, estimated
BF16 (full precision)1.2 GB2.7 GB
8-bit (Q8_0)0.64 GB2.1 GB
4-bit (Q4_K)0.34 GB1.8 GB
03

Architecture at a glance

SettingValue in config.json
Architecture classM2M100ForConditionalGeneration
Model typem2m_100
Layers12
Vocabulary256,206
Configured context1,024
Stored precisionfloat32
04

Will it fit on ANVE?

These verdicts are arithmetic, not tests, against the specification at launch: ANVE One can give a model about three quarters of its 64 GB of unified memory, while ANVE Office and ANVE Office Pro hold it on a 24 GB graphics card and keep the rest of the stack in 64 GB or 128 GB of system memory. ANVE’s tested default is Qwen 3.8 27B; this model is not yet validated on ANVE.

Configuration (at launch)Memory for the modelVerdict
ANVE Oneabout 48 GB of 64 GB unified memoryfits at BF16
ANVE Office24 GB on the RTX PRO 4000 Blackwell SFF Editionfits at BF16
ANVE Office Pro (GPU)24 GB on the RTX PRO 4000 Blackwellfits at BF16
05

What an office would use it for

For an office, offline translation keeps client documents on your own hardware. Have a fluent person check anything that will be filed or sent.

06

Variants with the same size

  • nllb-200-distilled-1.3B: a distilled, smaller-step version (same parameter count, so the memory figures above apply).
  • nllb-200-3.3B: a variant with the same size and memory needs (same parameter count, so the memory figures above apply).
  • nllb-200-1.3B: a variant with the same size and memory needs (same parameter count, so the memory figures above apply).
07

Similar models to compare

  • Hy-MT2-1.8B (2.0B parameters, about 2.6 GB at 4-bit before context; commercial use allowed)
08

Questions people ask

How much memory does nllb-200-distilled-600M need?
About 1.8 GB at 4-bit, 2.1 GB at 8-bit and 2.7 GB at full 16-bit precision, including about 1.5 GB of runtime overhead. These are estimates from its parameter count.
Can nllb-200-distilled-600M run on a 24 GB graphics card?
Yes, at BF16 by this estimate, with room left for context.
Is nllb-200-distilled-600M free for commercial use?
Its card lists CC BY-NC 4.0, which allows non-commercial use only, so an office cannot use it for paid client work without a separate licence. Read the licence itself before relying on this summary; it is not legal advice.
What is nllb-200-distilled-600M’s context window?
Its configuration sets 1,024 tokens. Longer contexts need more memory for the KV cache, so the practical limit on a given machine can be lower.
09

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.