Calculate the TPS of the Radeon RX 7900 XT on local AI models

AMD 20 GB GDDR6 800 GB/s November 2022

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

495 of 679 models it can run

Largest model it holds

Qwen3-Omni-30B-A3B

35.3B · IQ4_XS · 102 tok/s

Fastest model

Gemma 3 QAT 1B

264 tok/s · 1B

What AI models can a Radeon RX 7900 XT run?

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.

495 models match

Calculating
Quantisation Fit
264 tok/s

159–423 · low confidence

Gemma 3 1B 1B Mar 2025 1.8 GB 33k tokens Q8_0 Comfortable
264 tok/s

159–423 · low confidence

Gemma 3 QAT 1B 1B Apr 2025 1.8 GB 33k tokens Q8_0 Comfortable
264 tok/s

159–423 · low confidence

HGRN 1B (WT 103) 1B Nov 2023 1.8 GB 131k tokens ? Q8_0 Comfortable
264 tok/s

159–423 · low confidence

LLama 3..2 Typhoon 2 1B 1B Dec 2024 1.8 GB 131k tokens ? Q8_0 Comfortable
264 tok/s

159–423 · low confidence

OLMo-1B 1B Feb 2024 1.8 GB 131k tokens ? Q8_0 Comfortable
264 tok/s

159–423 · low confidence

Pythia-1b 1B Apr 2023 1.8 GB 131k tokens ? Q8_0 Comfortable
245 tok/s

147–392 · low confidence

OpenELM-1.1B 1.1B May 2024 1.9 GB 131k tokens ? Q8_0 Comfortable
240 tok/s

144–384 · low confidence

DeciCoder-1B 1.1B Aug 2023 1.9 GB 131k tokens ? Q8_0 Comfortable
240 tok/s

144–384 · low confidence

SantaCoder 1.1B Jan 2023 1.9 GB 131k tokens ? Q8_0 Comfortable
240 tok/s

144–384 · low confidence

TinyLlama-1.1B (1T token checkpoint) 1.1B Oct 2023 1.9 GB 131k tokens ? Q8_0 Comfortable
240 tok/s

144–384 · low confidence

TinyLlama-1.1B (3T token checkpoint) 1.1B Oct 2023 1.9 GB 131k tokens ? Q8_0 Comfortable
220 tok/s

132–352 · low confidence

EXAONE 4.0 (1.2B) 1.2B Jul 2025 2.0 GB 131k tokens ? Q8_0 Comfortable
220 tok/s

132–352 · low confidence

MinerU2.5 1.2B Sep 2025 2.0 GB 131k tokens ? Q8_0 Comfortable
220 tok/s

132–352 · low confidence

Pleias 1.0 1.2B 1.2B Dec 2024 2.0 GB 131k tokens ? Q8_0 Comfortable
220 tok/s

132–352 · low confidence

Pleias-RAG-1B 1.2B Apr 2025 2.0 GB 131k tokens ? Q8_0 Comfortable
215 tok/s

129–344 · low confidence

Llama 3.2 1B 1.2B Sep 2024 2.2 GB 131k tokens Q8_0 Comfortable
212 tok/s

127–339 · low confidence

MiniCPM-1.2B 1.2B Jun 2024 2.0 GB 131k tokens ? Q8_0 Comfortable
203 tok/s

122–325 · low confidence

DeepSeek Coder 1.3B 1.3B Jan 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
203 tok/s

122–325 · low confidence

DeepSeek-VL-1.3B 1.3B Mar 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
203 tok/s

122–325 · low confidence

DigiRL 1.3B Jun 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
203 tok/s

122–325 · low confidence

GLA Transformer 1.3B 1.3B Aug 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
203 tok/s

122–325 · low confidence

Janus 1.3B 1.3B Oct 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
203 tok/s

122–325 · low confidence

Kosmos-2.5 1.3B Aug 2024 2.1 GB 131k tokens ? Q8_0 Comfortable
203 tok/s

122–325 · low confidence

Otter 1.3B May 2023 2.1 GB 131k tokens ? Q8_0 Comfortable
203 tok/s

122–325 · low confidence

Phi-1 1.3B Oct 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

Radeon RX 7900 XT 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
20 GB
Memory bandwidth
800 GB/s
Memory type
GDDR6
Memory bus width
320 bit
Memory clock
2.5 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
Navi 31
Architecture
RDNA 3.0
Generation
Navi III(RX 7000)
Foundry
TSMC
Process size
5 nm
Transistors
57.7 billion
Transistor density
109,100 K/mm²
Die size
529 mm²
Package
MCM
Released
3 November 2022

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
1.39 GHz
Boost clock
2.39 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
5,376
Texture mapping units
336
Render output units
192
Ray tracing cores
84
L1 cache
250 KB
L2 cache
6 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)
103 TFLOPS
Single precision (FP32)
51.5 TFLOPS
Double precision (FP64)
1.6 TFLOPS
Pixel rate
460 GPixel/s
Texture rate
804 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)
300 W
Suggested power supply
700 W
Power connectors
2x 8-pin
Bus interface
PCIe 4.0 x16
Slot width
Dual-slot
Dimensions
276 mm × 51 mm
Display outputs
1x HDMI 2.1a, 2x DisplayPort 2.1, 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
2.2
Shader model
6.8

Listings

Where to buy a Radeon RX 7900 XT

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

20 GB

Bandwidth

800 GB/s

Largest model

Qwen3-Omni-30B-A3B

20 GB of GDDR6 puts the Radeon RX 7900 XT comfortably into small and mid-sized models, with roughly 18 GB usable once the driver overhead is taken out. The largest models are out of reach without splitting them.

Bandwidth is 800 GB/s across a 320-bit bus. Generating a token means reading every weight once, so that figure sets the pace more than any other number here, and at this level text arrives faster than most people read.

Bandwidth is clock times bus width, and this card clocks its memory at 2.5 GHz. Both halves matter, and neither is visible in a gaming benchmark.

Put together, the largest model that fits is Qwen3-Omni-30B-A3B at 35.3B, running IQ4_XS and producing around 102 tokens per second.

The chip and how it was built

The Radeon RX 7900 XT is built on the Navi 31 graphics processor, using AMD's RDNA 3.0 architecture, as part of the Navi III(RX 7000) generation.

The chip is manufactured by TSMC, on a 5 nm process, with a die measuring 529 mm², holding 57.7 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 November 2022, roughly 3 years ago. 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

103 TFLOPS

FP64

1.6 TFLOPS

On paper the Radeon RX 7900 XT reaches 103 TFLOPS at half precision and 51.5 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 is 1.6 TFLOPS. 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 1.39 GHz at base to 2.39 GHz boosted. 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

The Radeon RX 7900 XT has 250 KB of L1 cache, backed by 6 MB of L2. 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 5,376 shading units, 336 texture mapping units, and 192 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

300 W

The Radeon RX 7900 XT is rated at 300 W, with a 700 W power supply suggested for the whole system. 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 board occupies a dual-slot, measuring 276 mm long, and needs 2x 8-pin. Worth checking against the case and power supply already in the machine, since the largest cards need considerably more of both than a typical desktop provides.

It connects over PCIe 4.0 x16. The interface governs how quickly a model is loaded from disk into the card, not how fast it runs once there, so a narrower link costs a few seconds at startup and nothing thereafter.

The extremes

The largest AI models a Radeon RX 7900 XT can run

The biggest open-weight models that fit on this card, newest first. Each is shown at the best compression the card can hold.

  1. 01 Qwen3-Omni-30B-A3B 35.3B · IQ4_XS · Sep 2025 102 tok/s
  2. 02 Falcon-H1 34B · Q3_K_M · May 2025 21.0 tok/s
  3. 03 TeleChat2-35B 35B · Q3_K_M · Oct 2024 20.4 tok/s
  4. 04 Oryx 34B 34B · Q3_K_M · Sep 2024 21.0 tok/s
  5. 05 Smaug-34B 34B · Q3_K_M · Jul 2024 21.0 tok/s
  6. 06 Cambrian-1-34B 34B · Q3_K_M · Jun 2024 21.0 tok/s
  7. 07 Yi-1.5-34B 34B · Q3_K_M · May 2024 21.0 tok/s
  8. 08 CausalLM 34B β 34.4B · Q3_K_M · Feb 2024 20.7 tok/s
  9. 09 LLaVA-NeXT-34B (LLaVA-1.6) 34.8B · Q3_K_M · Jan 2024 20.5 tok/s
  10. 10 Poro 34B 34.2B · Q3_K_M · Dec 2023 20.9 tok/s

The fastest AI models on a Radeon RX 7900 XT

Where this card produces tokens quickest. Smaller models dominate here, because generating each token means reading the whole model out of memory once.

  1. 01 Gemma 3 QAT 1B 1B · Q8_0 · 1.8 GB 264 tok/s
  2. 02 Gemma 3 1B 1B · Q8_0 · 1.8 GB 264 tok/s
  3. 03 LLama 3..2 Typhoon 2 1B 1B · Q8_0 · 1.8 GB 264 tok/s
  4. 04 OLMo-1B 1B · Q8_0 · 1.8 GB 264 tok/s
  5. 05 HGRN 1B (WT 103) 1B · Q8_0 · 1.8 GB 264 tok/s
  6. 06 Pythia-1b 1B · Q8_0 · 1.8 GB 264 tok/s
  7. 07 OpenELM-1.1B 1.1B · Q8_0 · 1.9 GB 245 tok/s
  8. 08 TinyLlama-1.1B (1T token checkpoint) 1.1B · Q8_0 · 1.9 GB 240 tok/s
  9. 09 TinyLlama-1.1B (3T token checkpoint) 1.1B · Q8_0 · 1.9 GB 240 tok/s
  10. 10 DeciCoder-1B 1.1B · Q8_0 · 1.9 GB 240 tok/s

Step by step

How to work out the tokens per second of a Radeon RX 7900 XT

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.

  1. 01

    Find the model in the table

    Every one of the 495 models this Radeon RX 7900 XT runs is in the table above. Search narrows it by name or by size.

  2. 02

    Set the context length you will actually use

    Set the context to your real working length. Short questions cost almost nothing; a long document can consume a large share of the card's 20 GB.

  3. 03

    Set a minimum quality if you need one

    By default the table picks the least-compressed copy that fits. Setting a floor removes models that only qualify through heavy compression.

  4. 04

    Look at the range, not just the number

    The figures are calculated, not measured. 264 tok/s on Gemma 3 QAT 1B is the fastest result on this card, and like every row it carries a range that reflects how much the runtime matters.

  5. 05

    Check the headroom before you decide

    Compare what each model needs with the 20 GB this card provides. Tight means it works today; comfortable means it still works when the conversation grows.

  6. 06

    Open the model to compare cards

    Following a model through to its own page lists all the hardware that can run it, so you can see where the Radeon RX 7900 XT sits against the alternatives.

Answers

Radeon RX 7900 XT — common questions

01

What AI models can a Radeon RX 7900 XT run?

495 of the 679 open-weight language models we track fit on a Radeon RX 7900 XT and can be run locally on it. The table on this page lists every one, with the memory it needs, the quantisation it runs at and an estimated generation speed.

02

What is the largest AI model a Radeon RX 7900 XT can run?

The largest model in our catalogue that fits on a Radeon RX 7900 XT is Qwen3-Omni-30B-A3B at 35.3B parameters, compressed to IQ4_XS. It generates roughly 102 tokens per second and needs about 17.9 GB of the card's memory.

03

How many tokens per second does a Radeon RX 7900 XT produce?

It depends on the model. On a Radeon RX 7900 XT the fastest model we track is Gemma 3 QAT 1B at about 264 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.

04

Can a Radeon RX 7900 XT run a 7B model?

Yes. For example a Radeon RX 7900 XT runs Multi-Token Prediction 7B at Q8_0, using about 7.9 GB of memory and generating around 39.5 tokens per second.

05

Can a Radeon RX 7900 XT run a 13B model?

Yes. For example a Radeon RX 7900 XT runs DeepSeekMoE-16B at Q8_0, using about 17.5 GB of memory and generating around 91.8 tokens per second.

06

Can a Radeon RX 7900 XT run a 30B model?

Yes. For example a Radeon RX 7900 XT runs ERNIE-4.5-VL-28B-A3B at Q4_K_M, using about 16.2 GB of memory and generating around 121 tokens per second.

07

How much memory does a Radeon RX 7900 XT have?

A Radeon RX 7900 XT has 20 GB of GDDR6 memory. Around a tenth of that is reserved by the inference runtime and the driver, leaving roughly 18 GB available for a model and its conversation.

08

What is the memory bandwidth of a Radeon RX 7900 XT?

The Radeon RX 7900 XT has 800 GB/s of memory bandwidth, across a 320-bit memory bus. 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.

09

What type of memory does a Radeon RX 7900 XT use?

It uses GDDR6 clocked at 2.5 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.

10

Who makes the Radeon RX 7900 XT?

The Radeon RX 7900 XT is a AMD product, with the chip manufactured by TSMC, on a 5 nm process.

11

When was the Radeon RX 7900 XT released?

The Radeon RX 7900 XT was released in November 2022.

12

How much power does a Radeon RX 7900 XT use?

The Radeon RX 7900 XT has a rated board power of 300 W, and a 700 W system power supply is suggested. Generating text draws hard in bursts and idles between requests, so average consumption over a working session is normally well below the rated figure.

13

How much cache does a Radeon RX 7900 XT have?

The Radeon RX 7900 XT has 250 KB of L1 cache, and 6 MB of L2 cache. 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.

14

What are the TFLOPS of a Radeon RX 7900 XT?

The Radeon RX 7900 XT is rated at 103 TFLOPS at half precision and 51.5 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.

15

Does the Radeon RX 7900 XT support CUDA?

No. CUDA is NVIDIA-only, and the Radeon RX 7900 XT is a AMD card. 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.

16

What bus interface does the Radeon RX 7900 XT use?

It uses PCIe 4.0 x16. This governs how fast a model is loaded onto the card rather than how fast it runs once loaded, so it costs a few seconds at startup and nothing during generation.

17

Is the Radeon RX 7900 XT good for running local AI models?

Its memory covers small and mid-sized models, though the largest are out of reach and its bandwidth is high enough to generate text faster than most people read. In total it runs 495 of the models we track. Whether that is enough depends entirely on which model you want — the table above answers that directly.

18

Can a Radeon RX 7900 XT run a model that does not fit in its memory?

Only partly. Layers beyond the 20 GB sit in system memory and run at a fraction of the speed, so a mostly-offloaded model is rarely worth using. Every figure here assumes it is fully resident on the card.

19

Would two Radeon RX 7900 XT cards be twice as fast?

No. A second Radeon RX 7900 XT doubles the memory to 40 GB, which lets you hold models neither could hold alone, but generation does not split that way. These figures describe one card.

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.

All GPUs