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

ANVE

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

Voxtral-Mini-4B-Realtime-2602
The memory it needs, what you may do with it, and the hardware it suits.

Short answer. Voxtral-Mini-4B-Realtime-2602 needs about 4.0 GB of memory at 4-bit, 6.2 GB at 8-bit and 10 GB at full 16-bit precision, including runtime overhead. It has about 4.4 billion parameters and a 131,072-token context window. Licence in short: commercial use allowed. 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

Voxtral-Mini-4B-Realtime-2602 is a speech-recognition model that turns recorded speech into text, published by Mistral AI on Hugging Face on 21 January 2026. It has about 4.4 billion parameters and a configured context window of 131,072 tokens.

Its card lists the licence as Apache 2.0. In short, the licence allows commercial use, modification and redistribution, with attribution and a copy of the licence, and includes a patent grant.

Hugging Face counted about 1,921,051 downloads of it in the 30 days before we checked (7 Oct 2026), and 1,004 people have liked it there.

The card lists 13 languages: en, fr, es, de, ru, zh, ja, it, pt, nl, ar, hi and 1 more (ISO codes as given on the card).

It is built on mistralai/Ministral-3-3B-Base-2512, according to its 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.

Speech models also need working memory for the audio being processed, which is small next to the weights.

PrecisionWeightsTotal, estimated
BF16 (full precision)8.9 GB10 GB
8-bit (Q8_0)4.7 GB6.2 GB
4-bit (Q4_K)2.5 GB4.0 GB
03

Architecture at a glance

SettingValue in config.json
Architecture classVoxtralRealtimeForConditionalGeneration
Model typevoxtral_realtime
Layers26
Hidden size3,072
Attention heads32
Key-value heads8 (grouped-query attention)
Head size128
Vocabulary131,072
Configured context131,072
04

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 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, this is the model that turns a recorded meeting or call into a transcript. Accuracy depends heavily on the language, the microphone and how people talk over each other, so test it on your own recordings.

06

Similar models to compare

  • Phi-4-multimodal-instruct (5.6B parameters, about 4.6 GB at 4-bit before context; commercial use allowed)
  • Qwen3-ASR-1.7B (2.3B parameters, about 2.8 GB at 4-bit before context; commercial use allowed)
  • granite-4.0-1b-speech (2.3B parameters, about 2.8 GB at 4-bit before context; commercial use allowed)
07

Questions people ask

How much memory does Voxtral-Mini-4B-Realtime-2602 need?
About 4.0 GB at 4-bit, 6.2 GB at 8-bit and 10 GB at full 16-bit precision, including about 1.5 GB of runtime overhead. These are estimates from its parameter count.
Can Voxtral-Mini-4B-Realtime-2602 run on a 24 GB graphics card?
Yes, at BF16 by this estimate, with room left for context.
Is Voxtral-Mini-4B-Realtime-2602 free for commercial use?
Its card lists Apache 2.0, which allows commercial use, modification and redistribution, with attribution and a copy of the licence, and includes a patent grant. Read the licence itself before relying on this summary; it is not legal advice.
What is Voxtral-Mini-4B-Realtime-2602’s context window?
Its configuration sets 131,072 tokens. Longer contexts need more memory for the KV cache, so the practical limit on a given machine can be lower.
08

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.