Calculate the TPS of the FirePro W9000 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
Qwen-VL
9.6B · Q3_K_M · 24.5 tok/s
Fastest model
Gemma 3 QAT 1B
87.2 tok/s · 1B
Which AI models can run on a FirePro W9000?
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.
280 models match
Calculating| Quantisation | Fit | ||||||
|---|---|---|---|---|---|---|---|
|
87.2
tok/s
52–140 · low confidence |
Gemma 3 1B | 1B | Mar 2025 | 1.8 GB | 33k tokens | Q8_0 | Comfortable |
|
87.2
tok/s
52–140 · low confidence |
Gemma 3 QAT 1B | 1B | Apr 2025 | 1.8 GB | 33k tokens | Q8_0 | Comfortable |
|
87.2
tok/s
52–140 · low confidence |
HGRN 1B (WT 103) ≈ | 1B | Nov 2023 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
87.2
tok/s
52–140 · low confidence |
LLama 3..2 Typhoon 2 1B ≈ | 1B | Dec 2024 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
87.2
tok/s
52–140 · low confidence |
OLMo-1B ≈ | 1B | Feb 2024 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
87.2
tok/s
52–140 · low confidence |
Pythia-1b ≈ | 1B | Apr 2023 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
80.8
tok/s
48–129 · low confidence |
OpenELM-1.1B ≈ | 1.1B | May 2024 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
79.3
tok/s
48–127 · low confidence |
DeciCoder-1B ≈ | 1.1B | Aug 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
79.3
tok/s
48–127 · low confidence |
SantaCoder ≈ | 1.1B | Jan 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
79.3
tok/s
48–127 · low confidence |
TinyLlama-1.1B (1T token checkpoint) ≈ | 1.1B | Oct 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
79.3
tok/s
48–127 · low confidence |
TinyLlama-1.1B (3T token checkpoint) ≈ | 1.1B | Oct 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
72.7
tok/s
44–116 · low confidence |
EXAONE 4.0 (1.2B) ≈ | 1.2B | Jul 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
72.7
tok/s
44–116 · low confidence |
LFM2-1.2B ≈ | 1.2B | Jul 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
72.7
tok/s
44–116 · low confidence |
MinerU2.5 ≈ | 1.2B | Sep 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
72.7
tok/s
44–116 · low confidence |
Pleias 1.0 1.2B ≈ | 1.2B | Dec 2024 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
72.7
tok/s
44–116 · low confidence |
Pleias-RAG-1B ≈ | 1.2B | Apr 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
70.9
tok/s
43–113 · low confidence |
Llama 3.2 1B | 1.2B | Sep 2024 | 2.2 GB | 113k tokens | Q8_0 | Comfortable |
|
69.9
tok/s
42–112 · low confidence |
MiniCPM-1.2B ≈ | 1.2B | Jun 2024 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
67.1
tok/s
40–107 · low confidence |
DeepSeek Coder 1.3B ≈ | 1.3B | Jan 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
67.1
tok/s
40–107 · low confidence |
DeepSeek-VL-1.3B ≈ | 1.3B | Mar 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
67.1
tok/s
40–107 · low confidence |
DigiRL ≈ | 1.3B | Jun 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
67.1
tok/s
40–107 · low confidence |
GLA Transformer 1.3B ≈ | 1.3B | Aug 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
67.1
tok/s
40–107 · low confidence |
Janus 1.3B ≈ | 1.3B | Oct 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
67.1
tok/s
40–107 · low confidence |
Kosmos-2.5 ≈ | 1.3B | Aug 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
67.1
tok/s
40–107 · 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
FirePro W9000 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
- 6 GB
- Memory bandwidth
- 264 GB/s
- Memory type
- GDDR5
- Memory bus width
- 384 bit
- Memory clock
- 1.38 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
- Tahiti
- Architecture
- GCN 1.0
- Generation
- FirePro GCN(Wx000)
- Foundry
- TSMC
- Process size
- 28 nm
- Transistors
- 4.3 billion
- Transistor density
- 12,300 K/mm²
- Die size
- 352 mm²
- Released
- 14 June 2012
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
- 975 MHz
- Boost clock
- 975 MHz
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
- 2,048
- Texture mapping units
- 128
- Render output units
- 32
- L1 cache
- 16 KB
- L2 cache
- 0.75 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.
- Single precision (FP32)
- 4 TFLOPS
- Double precision (FP64)
- 998.4 GFLOPS
- Pixel rate
- 31 GPixel/s
- Texture rate
- 125 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)
- 274 W
- Suggested power supply
- 600 W
- Power connectors
- 1x 6-pin + 1x 8-pin
- Bus interface
- PCIe 3.0 x16
- Slot width
- Dual-slot
- Dimensions
- 279 mm
- Display outputs
- 6x mini-DisplayPort 1.2, 1x SDI
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
- 11.1
- OpenGL
- 4.6
- Vulkan
- 1.2
- OpenCL
- 1.2
- Shader model
- 5.1
Listings
Where to buy a FirePro W9000
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
Why memory is the number that matters here
Memory
6 GB
Bandwidth
264 GB/s
Largest model
Qwen-VL
FirePro W9000 carries only 6 GB of GDDR5. That limits it to the smaller end of the catalogue, and a model has to fit entirely inside before it generates anything at all. A runtime actually gets about 5.4 GB.
Memory bandwidth reaches 264 GB/s across a bus of 384 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.
Bandwidth is clock times bus width, and this card clocks its memory at 1.38 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.
The biggest thing it holds is Qwen-VL, 9.6B, compressed to Q3_K_M and generating around 24.5 tokens per second.
The chip and how it was built
FirePro W9000 is built on the graphics processor Tahiti, using the architecture GCN 1.0 from AMD, as part of the generation FirePro GCN(Wx000).
The chip is manufactured by TSMC, on a process of 28 nm, with a die measuring 352 mm², holding 4.3 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 June 2012, roughly 14.2512062136 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
FP64
998.4 GFLOPS
Double-precision throughput reaches 998.4 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 975 MHz to a boost of 975 MHz. 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
FirePro W9000 has an L1 cache of 16 KB, backed by an L2 cache of 0.75 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 2,048 shading units, 128 texture mapping units, and 32 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
274 W
FirePro W9000 is rated at 274 W, and the suggested system power supply is 600 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 board occupies dual-slot, measuring 279 mm long, and needs 1x 6-pin + 1x 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 3.0 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 that run on a FirePro W9000
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 FirePro W9000
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 FirePro W9000
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 280 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 6 GB it is often what pushes a large model over the edge.
-
03
Pin the comparison to one quality level
Each model is shown at the best compression this card can hold. A minimum quality hides the ones that only fit by being squeezed further than you would accept.
-
04
Take the range as the answer
Each speed is an estimate for a single conversation, with a range beneath it. The top end here is 87.2 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
Tight means it works today; comfortable means it still works when the conversation grows. Compare what each model needs against an available 6 GB.
-
06
Check the same model from the other side
Following a model through to its own page lists all the hardware that can run it, so you can see how it compares against FirePro W9000.
Answers
FirePro W9000 — common questions
FirePro W9000— can it run 7B models?
Yes. For example it runs Gemma 4 E4B at Q3_K_M, using about 5.2 GB of memory and generating around 52.3 tokens per second.
FirePro W9000— how much memory does it have?
This card has 6 GB of GDDR5. Around a tenth is reserved by the inference runtime and the driver, leaving roughly 5.4 GB available for a model and its conversation.
FirePro W9000— what is its memory bandwidth?
Memory bandwidth reaches 264 GB/s across a bus of 384 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.
FirePro W9000— what type of memory does it use?
It uses GDDR5 clocked at 1.38 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.
FirePro W9000— who makes it?
This is a product of AMD, with the chip manufactured by TSMC, on a process of 28 nm.
FirePro W9000— when was it released?
It was released in June 2012.
FirePro W9000— how much power does it use?
Rated board power is 274 W, and the suggested system power supply is 600 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.
FirePro W9000— how much cache does it have?
The L1 cache is 16 KB, and the L2 cache is 0.75 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.
FirePro W9000— 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.
FirePro W9000— what bus interface does it use?
It uses PCIe 3.0 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.
FirePro W9000— is it good for running local AI models?
Its memory limits it to smaller models though its bandwidth means generation will feel slow on larger models. In total it runs 280 of the models we track. Whether that is enough depends entirely on which model you want — the table above answers that directly.
FirePro W9000— can it run a model that does not fit in its memory?
Only partly. Layers beyond the card's 6 GB drags the whole thing down, and none of the figures on this page assume it.
Would two FirePro W9000 cards be twice as fast?
Capacity adds, throughput does not. Two of them give you 12 GB to work with rather than twice the tokens per second — every figure here is for a single card.
FirePro W9000— which AI models can it run?
280 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.
FirePro W9000— what is the largest AI model it can run?
The largest model in our catalogue that fits is Qwen-VL at 9.6B parameters, compressed to Q3_K_M. It generates roughly 24.5 tokens per second and needs about 5.4 GB of the card's memory.
FirePro W9000— 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 87.2 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.
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.