Hy3 Memory at each precision, the licence in plain words, and where it fits.
Short answer. Hy3 needs about 172 GB of memory at 4-bit, 322 GB at 8-bit and 602 GB at full 16-bit precision, including runtime overhead and an 8,192-token KV cache. It has about 298.8 billion parameters and a 262,144-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
Hy3 is a language model for chat, drafting and question answering, published by Tencent on Hugging Face on 2 July 2026. It has about 298.8 billion parameters and a configured context window of 262,144 tokens. It is a mixture-of-experts model: 192 experts, 8 used for each token.
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 407,420 downloads of it in the 30 days before we checked (7 Oct 2026), and 974 people have liked it there.
02
How much memory it needs
We size it the way you would size any model: the weights (parameters × bits ÷ 8), the KV cache for 8,192 tokens of context, and roughly 1.5 GB for the runtime itself. Real use varies with settings.
Each token of context adds about 0.33 MB of KV cache at 16-bit, so 32,768 tokens would add about 11 GB.
Precision
Weights
KV cache (8,192 tokens)
Total, estimated
BF16 (full precision)
598 GB
2.7 GB
602 GB
8-bit (Q8_0)
317 GB
2.7 GB
322 GB
4-bit (Q4_K)
168 GB
2.7 GB
172 GB
03
Architecture at a glance
Setting
Value in config.json
Architecture class
HYV3ForCausalLM
Model type
hy_v3
Layers
80
Hidden size
4,096
Attention heads
64
Key-value heads
8 (grouped-query attention)
Head size
128
Vocabulary
120,832
Experts
192, 8 active per token
Configured context
262,144
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-V4-Flash-0731 (304.2B parameters, about 173 GB at 4-bit before context; commercial use allowed)
DeepSeek-V4-Flash (290.9B parameters, about 165 GB at 4-bit before context; commercial use allowed)
MiMo-V2-Flash (309.8B parameters, about 176 GB at 4-bit before context; commercial use allowed)
07
Questions people ask
How much memory does Hy3 need?
About 172 GB at 4-bit, 322 GB at 8-bit and 602 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 Hy3 run on a 24 GB graphics card?
Not entirely: even at 4-bit it needs about 172 GB, so part of it would sit in system memory, which is much slower.
Is Hy3 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 Hy3’s context window?
Its configuration sets 262,144 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.
