Calculate the TPS of the Ryzen AI Z2 Extreme GPU on local AI models
Every model in our catalogue assessed against this card at the context length and minimum quality you choose. Speed is an estimate for a single request, calculated from this card's memory bandwidth and the size of each model once compressed.
Calculated for this card
721 models in our catalogue altogether
Largest model it holds
Nemotron 3-Nano-30B-A3B
31.6B · Q3_K_M · 40.1 tok/s
Fastest model
Gemma 3 QAT 1B
84.6 tok/s · 1B
Which AI models can run on a Ryzen AI Z2 Extreme GPU?
Set the inputs, read the answer
More context means more memory for the conversation cache. Speed is for a fresh conversation and does not change with this setting.
Hides models that would only fit by being compressed below this point.
455 models match
Calculating| Quantisation | Fit | ||||||
|---|---|---|---|---|---|---|---|
|
84.6
tok/s
51–135 · low confidence |
Gemma 3 1B | 1B | Mar 2025 | 1.8 GB | 33k tokens | Q8_0 | Comfortable |
|
84.6
tok/s
51–135 · low confidence |
Gemma 3 QAT 1B | 1B | Apr 2025 | 1.8 GB | 33k tokens | Q8_0 | Comfortable |
|
84.6
tok/s
51–135 · low confidence |
HGRN 1B (WT 103) ≈ | 1B | Nov 2023 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
84.6
tok/s
51–135 · low confidence |
LLama 3..2 Typhoon 2 1B ≈ | 1B | Dec 2024 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
84.6
tok/s
51–135 · low confidence |
OLMo-1B ≈ | 1B | Feb 2024 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
84.6
tok/s
51–135 · low confidence |
Pythia-1b ≈ | 1B | Apr 2023 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
78.3
tok/s
47–125 · low confidence |
OpenELM-1.1B ≈ | 1.1B | May 2024 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
76.9
tok/s
46–123 · low confidence |
DeciCoder-1B ≈ | 1.1B | Aug 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
76.9
tok/s
46–123 · low confidence |
SantaCoder ≈ | 1.1B | Jan 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
76.9
tok/s
46–123 · low confidence |
TinyLlama-1.1B (1T token checkpoint) ≈ | 1.1B | Oct 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
76.9
tok/s
46–123 · low confidence |
TinyLlama-1.1B (3T token checkpoint) ≈ | 1.1B | Oct 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
70.5
tok/s
42–113 · low confidence |
EXAONE 4.0 (1.2B) ≈ | 1.2B | Jul 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
70.5
tok/s
42–113 · low confidence |
LFM2-1.2B ≈ | 1.2B | Jul 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
70.5
tok/s
42–113 · low confidence |
MinerU2.5 ≈ | 1.2B | Sep 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
70.5
tok/s
42–113 · low confidence |
Pleias 1.0 1.2B ≈ | 1.2B | Dec 2024 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
70.5
tok/s
42–113 · low confidence |
Pleias-RAG-1B ≈ | 1.2B | Apr 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
68.8
tok/s
41–110 · low confidence |
Llama 3.2 1B | 1.2B | Sep 2024 | 2.2 GB | 131k tokens | Q8_0 | Comfortable |
|
67.8
tok/s
41–108 · low confidence |
MiniCPM-1.2B ≈ | 1.2B | Jun 2024 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
65.1
tok/s
39–104 · low confidence |
DeepSeek Coder 1.3B ≈ | 1.3B | Jan 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
65.1
tok/s
39–104 · low confidence |
DeepSeek-VL-1.3B ≈ | 1.3B | Mar 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
65.1
tok/s
39–104 · low confidence |
DigiRL ≈ | 1.3B | Jun 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
65.1
tok/s
39–104 · low confidence |
GLA Transformer 1.3B ≈ | 1.3B | Aug 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
65.1
tok/s
39–104 · low confidence |
Janus 1.3B ≈ | 1.3B | Oct 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
65.1
tok/s
39–104 · low confidence |
Kosmos-2.5 ≈ | 1.3B | Aug 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
65.1
tok/s
39–104 · low confidence |
Otter ≈ | 1.3B | May 2023 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
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
Ryzen AI Z2 Extreme GPU full specification
Everything on record for this board, ordered by how much it bears on running a language model rather than by how a spec sheet would list it. Memory comes first because it decides the outcome; the rest is context.
Memory
The two specifications that decide what this card can run and how quickly. Capacity sets which models fit; bandwidth sets how many tokens per second they produce once they do.
- Memory size
- 16 GB
- Memory bandwidth
- 256 GB/s
- Memory type
- LPDDR5X
- Memory bus width
- 256 bit
- Memory clock
- 1 GHz
The chip
Which processor is on the board and how it was manufactured. A smaller process size generally means more performance for the same power.
- Graphics processor
- Strix Point
- Architecture
- RDNA 3.5
- Generation
- Console GPU(AMD)
- Foundry
- TSMC
- Process size
- 4 nm
- Transistors
- 34 billion
- Transistor density
- 145,900 K/mm²
- Die size
- 233 mm²
- Package
- FP8
- Released
- 16 October 2025
Clock speeds
How fast the processor runs. Worth far less here than on a gaming benchmark: generating text is limited by memory bandwidth, so a higher clock barely moves the result.
- Base clock
- 800 MHz
- Boost clock
- 2.7 GHz
Processing units
What the chip contains. These drive graphics performance and matter mainly for processing a long prompt rather than for producing the answer.
- Shading units
- 1,024
- Texture mapping units
- 64
- Render output units
- 48
- Ray tracing cores
- 16
- L1 cache
- 128 KB
- L2 cache
- 8 MB
Theoretical performance
Peak arithmetic rates published for the board. These are ceilings that no real workload reaches, and generating text reaches a small fraction of them because it is limited by memory rather than arithmetic.
- Half precision (FP16)
- 11.1 TFLOPS
- Single precision (FP32)
- 5.5 TFLOPS
- Double precision (FP64)
- 345.6 GFLOPS
- Pixel rate
- 130 GPixel/s
- Texture rate
- 173 GTexel/s
The board
What it takes to physically install and power the card — the practical constraints that decide whether it fits the machine you already own.
- Power draw (TDP)
- 28 W
- Power connectors
- None
- Display outputs
- 1x USB Type-C
Software support
Which graphics and compute interfaces the card supports. CUDA compute capability is the one that bears on inference: below 7.0 there are no tensor cores, and modern inference software falls back to slower code paths.
- DirectX
- 12.2
- OpenGL
- 4.6
- Vulkan
- 1.4
- OpenCL
- 2.1
- Shader model
- 6.8
Listings
Where to buy a Ryzen AI Z2 Extreme GPU
No vendor is currently listing this card. Listings come from vendors who publish them here directly — browse the vendor directory to see who is selling what.
What the numbers mean
Capacity and bandwidth
Memory
16 GB
Bandwidth
256 GB/s
Largest model
Nemotron 3-Nano-30B-A3B
Ryzen AI Z2 Extreme GPU carries 16 GB of LPDDR5X. That reaches comfortably into small and mid-sized models, though the largest stay out of reach without splitting them. Driver overhead leaves roughly 14.4 GB.
Memory bandwidth reaches 256 GB/s across a bus of 256 bits. Bandwidth is this card's real constraint. Every token requires reading the entire model out of memory, so a large model will feel slow here even when it fits.
That comes from a memory clock of 1 GHz. Widening the bus and raising the clock are the two levers a manufacturer has, which is why a card with unremarkable cores can still generate quickly.
In practice that combination tops out at Nemotron 3-Nano-30B-A3B, 31.6B, compressed to Q3_K_M and generating around 40.1 tokens per second.
The chip and how it was built
Ryzen AI Z2 Extreme GPU is built on the graphics processor Strix Point, using the architecture RDNA 3.5 from AMD, as part of the generation Console GPU(AMD).
The chip is manufactured by TSMC, on a process of 4 nm, with a die measuring 233 mm², holding 34 billion transistors. A smaller process generally means more performance for the same power, though for language models it matters far less than the memory subsystem.
It was released in October 2025. Inference software support tends to follow hardware by a year or two, so a card of this age generally has mature, well-optimised code paths available to it.
Compute throughput, and why it matters less than it looks
FP16
11.1 TFLOPS
FP64
345.6 GFLOPS
On paper Ryzen AI Z2 Extreme GPU reaches 11.1 TFLOPS at half precision, and 5.5 TFLOPS at single precision. These are peak figures no real workload sustains, and generating text reaches only a small fraction of them — decoding is limited by memory rather than arithmetic, which is why a card can look enormously powerful here and still produce tokens at an ordinary rate.
Double-precision throughput reaches 345.6 GFLOPS. It has no bearing on running a language model — no inference runtime uses it — but it separates datacentre parts from consumer ones, since the latter deliberately restrict it.
Clocks run from a base of 800 MHz to a boost of 2.7 GHz. Worth far less here than on a gaming benchmark: raising the clock speeds up the arithmetic, and the arithmetic is not what generation is waiting on.
Cache and processing units
Ryzen AI Z2 Extreme GPU has an L1 cache of 128 KB, backed by an L2 cache of 8 MB. Cache absorbs a share of the memory traffic that would otherwise hit the main bus, which is the one place on this page where a number other than bandwidth quietly affects generation speed — a large L2 lets more of the working set stay close to the cores.
There are 1,024 shading units, 64 texture mapping units, and 48 render output units. These drive graphics workloads and contribute to prompt processing, but they sit idle for much of the time a model spends generating a reply.
Power, size and installation
Power draw
28 W
Ryzen AI Z2 Extreme GPU is rated at 28 W. Running a language model keeps a card busy in bursts rather than continuously — it draws hard while generating and idles between requests — so sustained draw over a working day is usually well below the rated figure.
The extremes
The largest AI models that run on a Ryzen AI Z2 Extreme GPU
The biggest open-weight models that fit on this card, newest first. Each is shown at the best compression the card can hold.
The fastest AI models on a Ryzen AI Z2 Extreme GPU
Where this card produces tokens quickest. Smaller models dominate here, because generating each token means reading the whole model out of memory once.
Step by step
How to work out the tokens per second of a Ryzen AI Z2 Extreme GPU
You do not have to calculate anything by hand — the gputps.com calculator on this page has already worked it out for every model this card can hold. Reading off the answer takes six steps.
-
01
Start with the model, not the specification
The table lists 455 models the card handles. The search box takes a name or a size such as 27b, which matches on parameter count.
-
02
Match the context to your work
Drag the slider to the conversation length you plan to work at. The cache grows with the conversation, and against a card holding 16 GB that is frequently the difference between a model fitting and not.
-
03
Pin the comparison to one quality level
Compression is what lets bigger models fit. The quality control drops any model that needs more of it than you are willing to give.
-
04
Read the speed and the range
Speeds come with error bars for a reason. The best case here is 84.6 tok/s on Gemma 3 QAT 1B. The same card and model vary by thirty to fifty per cent between inference runtimes.
-
05
Check the headroom before you decide
The fit column separates models that just fit from those with room to spare — worth checking before settling on one, against an available 16 GB.
-
06
Cross-check against other hardware
Following a model through to its own page lists all the hardware that can run it, so you can see how it compares against Ryzen AI Z2 Extreme GPU.
Answers
Ryzen AI Z2 Extreme GPU — common questions
Ryzen AI Z2 Extreme GPU— does it support CUDA?
No. CUDA is NVIDIA-only, and this is a card from AMD. It runs language models through ROCm, Vulkan or Metal depending on the software, which are less mature than the CUDA path — our estimates apply a penalty for that.
Ryzen AI Z2 Extreme GPU— is it good for running local AI models?
Its memory covers small and mid-sized models, though the largest are out of reach though its bandwidth means generation will feel slow on larger models. In total it runs 455 of the models we track. Whether that is enough depends entirely on which model you want — the table above answers that directly.
Ryzen AI Z2 Extreme GPU— can it run a model that does not fit in its memory?
Only partly. Layers beyond the card's 16 GB is possible and usually a false economy: the system-memory portion is slow enough to dominate the result.
Would two Ryzen AI Z2 Extreme GPU cards be twice as fast?
No. A second card doubles the memory to 32 GB which lets you hold models neither could hold alone, but generation does not split that way. These figures describe one card.
Ryzen AI Z2 Extreme GPU— which AI models can it run?
455 of the 721 open-weight language models we track fit on this card and can be run locally. The table on this page lists every one, with the memory it needs, the quantisation it runs at and an estimated generation speed.
Ryzen AI Z2 Extreme GPU— what is the largest AI model it can run?
The largest model in our catalogue that fits is Nemotron 3-Nano-30B-A3B at 31.6B parameters, compressed to Q3_K_M. It generates roughly 40.1 tokens per second and needs about 14.4 GB of the card's memory.
Ryzen AI Z2 Extreme GPU— how many tokens per second does it produce?
It depends on the model. The fastest model we track here is Gemma 3 QAT 1B at about 84.6 tokens per second, while larger models run proportionally slower because each token requires reading the whole model out of memory once. Speeds are estimates for a single conversation at a time.
Ryzen AI Z2 Extreme GPU— can it run 7B models?
Yes. For example it runs Gemma 4 E4B at Q8_0, using about 9.8 GB of memory and generating around 18.8 tokens per second.
Ryzen AI Z2 Extreme GPU— can it run 13B models?
Yes. For example it runs DeepSeekMoE-16B at Q6_K, using about 13.7 GB of memory and generating around 42.7 tokens per second.
Ryzen AI Z2 Extreme GPU— can it run 30B models?
Yes. For example it runs ERNIE-4.5-VL-28B-A3B at Q3_K_M, using about 12.9 GB of memory and generating around 45.3 tokens per second.
Ryzen AI Z2 Extreme GPU— how much memory does it have?
This card has 16 GB of LPDDR5X. Around a tenth is reserved by the inference runtime and the driver, leaving roughly 14.4 GB available for a model and its conversation.
Ryzen AI Z2 Extreme GPU— what is its memory bandwidth?
Memory bandwidth reaches 256 GB/s across a bus of 256 bits. This is the single best predictor of how fast it generates text, because producing each token means reading the entire model out of memory once.
Ryzen AI Z2 Extreme GPU— what type of memory does it use?
It uses LPDDR5X clocked at 1 GHz. HBM types are found on datacentre accelerators and carry far more bandwidth than the GDDR used on desktop cards, which is why they generate tokens considerably faster at the same capacity.
Ryzen AI Z2 Extreme GPU— who makes it?
This is a product of AMD, with the chip manufactured by TSMC, on a process of 4 nm.
Ryzen AI Z2 Extreme GPU— when was it released?
It was released in October 2025.
Ryzen AI Z2 Extreme GPU— how much power does it use?
Rated board power is 28 W. Generating text draws hard in bursts and idles between requests, so average consumption over a working session is normally well below the rated figure.
Ryzen AI Z2 Extreme GPU— how much cache does it have?
The L1 cache is 128 KB, and the L2 cache is 8 MB. Cache absorbs part of the memory traffic that would otherwise reach the main bus, so a larger L2 gives a modest lift to generation speed beyond what bandwidth alone predicts.
Ryzen AI Z2 Extreme GPU— what are its TFLOPS?
It is rated at 11.1 TFLOPS at half precision and 5.5 TFLOPS at single precision. These are peak arithmetic ceilings rather than achievable rates, and text generation reaches only a small fraction of them because it is limited by memory bandwidth instead.
The other direction
Looking at it from the other side?
This page starts from the hardware. If you already know which model you want and need to know what it takes to run it, start from the model instead.