Qwen3-235B-A22B The memory it needs, what you may do with it, and the hardware it suits.
Short answer. Qwen3-235B-A22B needs about 135 GB of memory at 4-bit, 253 GB at 8-bit and 473 GB at full 16-bit precision, including runtime overhead and an 8,192-token KV cache. It has about 235.1 billion parameters and a 40,960-token context window. Licence in short: commercial use allowed. By memory alone, on ANVE’s launch configurations: ANVE One does not fit, ANVE Office does not fit, ANVE Office Pro (GPU) does not fit. These are estimates from its published size, not benchmarks.
01
What it is
Qwen3-235B-A22B is a language model for chat, drafting and question answering, published by Alibaba’s Qwen team on Hugging Face on 27 April 2025. It has about 235.1 billion parameters and a configured context window of 40,960 tokens. It is a mixture-of-experts model: 128 experts, 8 used for each token, so only about 22B parameters are active at a time.
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 526,108 downloads of it in the 30 days before we checked (7 Oct 2026), and 1,111 people have liked it there.
02
How much memory it needs
How these figures are worked out: weights = parameters × bits per weight ÷ 8. The KV cache is added for 8,192 tokens of context, plus about 1.5 GB of runtime overhead. Treat the totals as planning estimates.
Each token of context adds about 0.19 MB of KV cache at 16-bit, so 32,768 tokens would add about 6.3 GB.
Precision
Weights
KV cache (8,192 tokens)
Total, estimated
BF16 (full precision)
470 GB
1.6 GB
473 GB
8-bit (Q8_0)
250 GB
1.6 GB
253 GB
4-bit (Q4_K)
132 GB
1.6 GB
135 GB
03
Architecture at a glance
Setting
Value in config.json
Architecture class
Qwen3MoeForCausalLM
Model type
qwen3_moe
Layers
94
Hidden size
4,096
Attention heads
64
Key-value heads
4 (grouped-query attention)
Head size
128
Vocabulary
151,936
Experts
128, 8 active per token
Configured context
40,960
Stored precision
bfloat16
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 model
Verdict
ANVE One
about 48 GB of 64 GB unified memory
does not fit
ANVE Office
24 GB on the RTX PRO 4000 Blackwell SFF Edition
does not fit
ANVE Office Pro (GPU)
24 GB on the RTX PRO 4000 Blackwell
does not fit
05
What an office would use it for
This is frontier-scale: even at 4-bit it needs far more memory than any single office box, so it belongs on a multi-GPU server or a cloud service, not on an appliance.
06
Similar models to compare
DeepSeek-V2 (235.7B parameters, about 134 GB at 4-bit before context; custom licence: check before commercial use)
MiniMax-M2.7 (228.7B parameters, about 130 GB at 4-bit before context; custom licence: check before commercial use)
DeepSeek-V4-Flash (290.9B parameters, about 165 GB at 4-bit before context; commercial use allowed)
07
Questions people ask
How much memory does Qwen3-235B-A22B need?
About 135 GB at 4-bit, 253 GB at 8-bit and 473 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 Qwen3-235B-A22B run on a 24 GB graphics card?
Not entirely: even at 4-bit it needs about 135 GB, so part of it would sit in system memory, which is much slower.
Is Qwen3-235B-A22B 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 Qwen3-235B-A22B’s context window?
Its configuration sets 40,960 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.
