Qwen3.5 397B-A17B TPS calculator
Each card below is assessed against this model at the context length and minimum quality you choose. Speed is an estimate for a single request, calculated from the card's memory bandwidth and the size of the model once compressed.
Calculated for this model
818 cards we hold specifications for
Smallest card that fits
Radeon Instinct MI300X
192 GB · Q3_K_M · 66.4 tok/s
Fastest card
B300
109 tok/s · 288 GB
Which GPUs can run Qwen3.5 397B-A17B?
Set the inputs, read the answer
A longer conversation needs more memory, which can push this model off smaller cards.
Hides cards that would only fit the model by compressing it below this point.
6 cards match
Calculating| Needs | Quantisation | Fit | |||||
|---|---|---|---|---|---|---|---|
|
109
tok/s
66–175 · low confidence |
B300 NVIDIA | 288 GB | 8,000 GB/s | Sep 2025 | 213.2 GB | Q4_K_M | Comfortable |
|
87.4
tok/s
52–140 · low confidence |
Radeon Instinct MI350X AMD | 288 GB | 8,190 GB/s | Jan 2025 | 213.2 GB | Q4_K_M | Comfortable |
|
87.4
tok/s
52–140 · low confidence |
Radeon Instinct MI355X AMD | 288 GB | 8,190 GB/s | Jan 2025 | 213.2 GB | Q4_K_M | Comfortable |
|
66.4
tok/s
40–106 · low confidence |
Radeon Instinct MI300X AMD | 192 GB | 5,325 GB/s | Dec 2023 | 167.0 GB | Q3_K_M | Tight |
|
66.4
tok/s
40–106 · low confidence |
Radeon Instinct MI308X AMD | 192 GB | 5,325 GB/s | Dec 2023 | 167.0 GB | Q3_K_M | Tight |
|
64.0
tok/s
38–102 · low confidence |
Radeon Instinct MI325X AMD | 256 GB | 6,000 GB/s | Oct 2024 | 213.2 GB | Q4_K_M | Tight |
Speeds are estimates for a single request — one conversation at a time — calculated from memory bandwidth, model size and quantisation. Real throughput varies with the inference runtime and its version. Figures published by hardware vendors measure many simultaneous requests and are much higher.
On record
Full specification
Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.
Origin
Who built this model, where, and when it was published.
- Organisation
- Alibaba
- Organisation type
- Industry
- Country
- China
- Published
- 13 February 2026
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language, Vision
- Task
- Language modeling/generation, Vision-language generation
Size
How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.
- Parameters
- 397B
- Training data
- tokens
"it comprises 397 billion total parameters"
Availability
Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.
- Weights
- Open — downloadable
- Model access
- Open weights (unrestricted)
- Hugging Face
- Qwen
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Why it is tracked
- Discretionary
- Record confidence
- Likely
Sources
Where this record came from and when it was last checked.
- Reference
- Qwen3.5: Towards Native Multimodal Agents
- Last updated
- 8 April 2026
The extremes
The ten fastest GPUs that run Qwen3.5 397B-A17B
Ranked by estimated tokens per second, newest card first where speeds tie. Because generation is bound by memory bandwidth, this ordering follows bandwidth rather than any gaming benchmark.
- 01 B300 288 GB · 8,000 GB/s · Q4_K_M 109 tok/s
- 02 Radeon Instinct MI350X 288 GB · 8,190 GB/s · Q4_K_M 87.4 tok/s
- 03 Radeon Instinct MI355X 288 GB · 8,190 GB/s · Q4_K_M 87.4 tok/s
- 04 Radeon Instinct MI300X 192 GB · 5,325 GB/s · Q3_K_M 66.4 tok/s
- 05 Radeon Instinct MI308X 192 GB · 5,325 GB/s · Q3_K_M 66.4 tok/s
- 06 Radeon Instinct MI325X 256 GB · 6,000 GB/s · Q4_K_M 64.0 tok/s
The smallest GPUs that still run Qwen3.5 397B-A17B
The cheapest route in, by memory capacity. A tight fit runs the model but leaves nothing spare for a longer conversation.
- 01 Radeon Instinct MI300X 192 GB · needs 167.0 GB · Q3_K_M · tight 66.4 tok/s
- 02 Radeon Instinct MI308X 192 GB · needs 167.0 GB · Q3_K_M · tight 66.4 tok/s
- 03 Radeon Instinct MI325X 256 GB · needs 213.2 GB · Q4_K_M · tight 64.0 tok/s
- 04 B300 288 GB · needs 213.2 GB · Q4_K_M · comfortable 109 tok/s
- 05 Radeon Instinct MI350X 288 GB · needs 213.2 GB · Q4_K_M · comfortable 87.4 tok/s
- 06 Radeon Instinct MI355X 288 GB · needs 213.2 GB · Q4_K_M · comfortable 87.4 tok/s
What the numbers mean
Hardware requirements in practice
Minimum card
Radeon Instinct MI300X
Memory needed
167.0 GB
Fastest
109 tok/s
Qwen3.5 397B-A17B reaches a parameter count of 397B. That is beyond what any single graphics card holds. Running it means either splitting it across several cards or renting hardware built for the job, and every card able to hold it alone is a datacentre part. The number that can: 6.
The entry point is Radeon Instinct MI300X, with a memory capacity of 192 GB, running it at a compression of Q3_K_M and producing around 66.4 tokens per second.
At the other end sits B300, generating roughly 109 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.
Background
Qwen3.5 397B-A17B was published by Alibaba, in the country recorded as China, during February 2026. It comes out of an organisation categorised as industry.
It works in the domain of Language, Vision, and is recorded as performing the task of language modeling/generation, Vision-language generation.
The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached. On Hugging Face it is published under the organisation Qwen.
Reading the throughput figures
Half the cards that hold it manage more than 76.9 tokens per second. Producing text faster than most people read it: 6 of them.
Mixture-of-experts routing is why the speeds here look high for the parameter count. Only a fraction is read per token, but the whole thing has to be loaded.
Memory here is estimated from size rather than computed from the architecture, which is not recorded for this model — the numbers are indicative rather than exact.
Training and provenance
Its inclusion criterion: discretionary.
Step by step
How to choose a GPU for Qwen3.5 397B-A17B
The table above has already assessed every card we hold specifications for against this model. Getting to your answer takes six steps.
-
01
Read the memory figure first
The table lists every card able to hold Qwen3.5 397B-A17B, needing around 167.0 GB at a compression of Q3_K_M. That figure, not the headline performance of a card, is what decides whether it runs.
-
02
Set the context length you will work at
The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect, because at long context a card that handles short questions easily can be dropped by Qwen3.5 397B-A17B.
-
03
Decide how much compression you will accept
Compression is what makes a model fit smaller cards, at some cost in accuracy, reaching a compression of Q3_K_M on the smallest card that fits. Setting a minimum quality drops the cards that only manage it by squeezing further than you would want, and holds the comparison at one level.
-
04
Sort by speed
Sort by speed to see how cards rank for Qwen3.5 397B-A17B. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B300, at 109 tok/s.
-
05
Check the fit verdict before buying
A tight fit runs, but leaves nothing spare for a longer conversation, in the case of Qwen3.5 397B-A17B. Comfortable means you can grow the context later. That difference matters more than a few tokens per second, so buy for comfortable if you expect to.
-
06
See what else that card runs
Each card page repeats this sweep for every model we hold, answering what else the hardware is good for beyond Qwen3.5 397B-A17B.
Answers
Qwen3.5 397B-A17B — common questions
Qwen3.5 397B-A17B— where can I download it?
Its weights are published on Hugging Face, under the organisation Qwen. We do not host model files — this site calculates what hardware is needed to run them.
Qwen3.5 397B-A17B— can I run it if it does not fit in my GPU?
It can be split between the card and system memory, but it generates painfully slowly that way. The nearest miss we calculate falls short by 51.2 GB. Every figure here assumes the whole model is resident on the card.
Qwen3.5 397B-A17B— would two GPUs run it faster?
A second card roughly doubles the memory available but not the generation rate. The number already able to run it alone: 6. So a second card is rarely the answer here.
Qwen3.5 397B-A17B— why does the quantisation differ between cards?
Because capacity varies, so does how hard it has to be squeezed. The number of distinct levels in the table above: 2. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.
Qwen3.5 397B-A17B— how accurate are these speed estimates?
They are calculated from specifications rather than measured, and each carries a range. One example: 66–175 tok/s on B300. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.
Qwen3.5 397B-A17B— what GPU do I need to run it?
The smallest card in our catalogue that holds it is Radeon Instinct MI300X, with a memory capacity of 192 GB. It runs the model at a compression of Q3_K_M using about 167.0 GB, and produces roughly 66.4 tokens per second. The number of cards able to run it in total: 6.
Qwen3.5 397B-A17B— how fast is it on a GPU?
It depends on the card. The quickest we calculate is B300, at about 109 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and the number of cards clearing that: 6.
Qwen3.5 397B-A17B— how much VRAM does it need?
It needs about 167.0 GB at a compression of Q3_K_M, which is what the smallest card that runs it uses. Less compression needs more: the figures in the memory column above are recalculated for each card, because each one holds the least-compressed version it can.
Qwen3.5 397B-A17B— is it open source?
Its weights are published, so it can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.
Qwen3.5 397B-A17B— how many parameters does it have?
It has a parameter count of 397B. "it comprises 397 billion total parameters". That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.
Qwen3.5 397B-A17B— who created it?
It was published by Alibaba, based in China, an organisation categorised as industry.
Qwen3.5 397B-A17B— when was it released?
It was published in February 2026.
Qwen3.5 397B-A17B— what is it used for?
It works in the domain of Language, Vision, and is recorded as handling the task of language modeling/generation, Vision-language generation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
The other direction
Looking at it from the other side?
This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.