mxbai-rerank-xsmall-v1 The memory it needs, what you may do with it, and the hardware it suits.
Short answer. mxbai-rerank-xsmall-v1 needs about 1.5 GB of memory at 4-bit, 1.6 GB at 8-bit and 1.6 GB at full 16-bit precision, including runtime overhead. It has about 71 million parameters and a 512-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
mxbai-rerank-xsmall-v1 is a reranker that re-orders search results so the most relevant passages come first, published by Mixedbread on Hugging Face on 29 February 2024. It has about 71 million parameters and a configured context window of 512 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 665,307 downloads of it in the 30 days before we checked (7 Oct 2026), and 59 people have liked it there.
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.
Precision
Weights
Total, estimated
BF16 (full precision)
0.14 GB
1.6 GB
8-bit (Q8_0)
0.08 GB
1.6 GB
4-bit (Q4_K)
0.04 GB
1.5 GB
03
Architecture at a glance
Setting
Value in config.json
Architecture class
DebertaV2ForSequenceClassification
Model type
deberta-v2
Layers
12
Hidden size
384
Attention heads
6
Head size
64
Vocabulary
128,100
Configured context
512
Stored precision
float16
04
Will it fit on ANVE?
Arithmetic against the launch specification, not a test: about three quarters of ANVE One’s 64 GB of unified memory for the model (as AMD’s Variable Graphics Memory allows), and the 24 GB on the graphics card in ANVE Office and ANVE Office Pro, which keep the rest of the stack in 64 GB and 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 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
05
What an office would use it for
A reranker sits after search: it reads the top results and the question together and puts the best passages first, which improves the answers a document assistant gives.
06
Similar models to compare
Qwen3-Reranker-0.6B (0.6B parameters, about 1.8 GB at 4-bit before context; commercial use allowed)
llama-nemotron-rerank-1b-v2 (1.2B parameters, about 2.2 GB at 4-bit before context; custom licence: check before commercial use)
Qwen3-VL-Reranker-2B (2.1B parameters, about 2.7 GB at 4-bit before context; commercial use allowed)
07
Questions people ask
How much memory does mxbai-rerank-xsmall-v1 need?
About 1.5 GB at 4-bit, 1.6 GB at 8-bit and 1.6 GB at full 16-bit precision, including about 1.5 GB of runtime overhead. These are estimates from its parameter count.
Can mxbai-rerank-xsmall-v1 run on a 24 GB graphics card?
Yes, at BF16 by this estimate, with room left for context.
Is mxbai-rerank-xsmall-v1 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 mxbai-rerank-xsmall-v1’s context window?
Its configuration sets 512 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.
