Calculate the TPS of the Tesla P40 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
679 models in our catalogue altogether
Largest model it holds
Mixtral 8x7B
46.7B · Q3_K_M · 26.1 tok/s
Fastest model
Gemma 3 QAT 1B
125 tok/s · 1B
Which AI models can run on a Tesla P40?
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.
502 models match
Calculating| Quantisation | Fit | ||||||
|---|---|---|---|---|---|---|---|
|
125
tok/s
44–250 · low confidence |
Gemma 3 1B | 1B | Mar 2025 | 1.8 GB | 33k tokens | Q8_0 | Comfortable |
|
125
tok/s
44–250 · low confidence |
Gemma 3 QAT 1B | 1B | Apr 2025 | 1.8 GB | 33k tokens | Q8_0 | Comfortable |
|
125
tok/s
44–250 · low confidence |
HGRN 1B (WT 103) ≈ | 1B | Nov 2023 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
125
tok/s
44–250 · low confidence |
LLama 3..2 Typhoon 2 1B ≈ | 1B | Dec 2024 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
125
tok/s
44–250 · low confidence |
OLMo-1B ≈ | 1B | Feb 2024 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
125
tok/s
44–250 · low confidence |
Pythia-1b ≈ | 1B | Apr 2023 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
116
tok/s
41–231 · low confidence |
OpenELM-1.1B ≈ | 1.1B | May 2024 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
114
tok/s
40–227 · low confidence |
DeciCoder-1B ≈ | 1.1B | Aug 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
114
tok/s
40–227 · low confidence |
SantaCoder ≈ | 1.1B | Jan 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
114
tok/s
40–227 · low confidence |
TinyLlama-1.1B (1T token checkpoint) ≈ | 1.1B | Oct 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
114
tok/s
40–227 · low confidence |
TinyLlama-1.1B (3T token checkpoint) ≈ | 1.1B | Oct 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
104
tok/s
36–208 · low confidence |
EXAONE 4.0 (1.2B) ≈ | 1.2B | Jul 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
104
tok/s
36–208 · low confidence |
MinerU2.5 ≈ | 1.2B | Sep 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
104
tok/s
36–208 · low confidence |
Pleias 1.0 1.2B ≈ | 1.2B | Dec 2024 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
104
tok/s
36–208 · low confidence |
Pleias-RAG-1B ≈ | 1.2B | Apr 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
102
tok/s
36–203 · low confidence |
Llama 3.2 1B | 1.2B | Sep 2024 | 2.2 GB | 131k tokens | Q8_0 | Comfortable |
|
100
tok/s
35–200 · low confidence |
MiniCPM-1.2B ≈ | 1.2B | Jun 2024 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
DeepSeek Coder 1.3B ≈ | 1.3B | Jan 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
DeepSeek-VL-1.3B ≈ | 1.3B | Mar 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
DigiRL ≈ | 1.3B | Jun 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
GLA Transformer 1.3B ≈ | 1.3B | Aug 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
Janus 1.3B ≈ | 1.3B | Oct 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
Kosmos-2.5 ≈ | 1.3B | Aug 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
Otter ≈ | 1.3B | May 2023 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · 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
Tesla P40 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
- 24 GB
- Memory bandwidth
- 347 GB/s
- Memory type
- GDDR5
- Memory bus width
- 384 bit
- Memory clock
- 1.81 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
- GP102
- Architecture
- Pascal
- Generation
- Tesla Pascal(Pxx)
- Foundry
- TSMC
- Process size
- 16 nm
- Transistors
- 11.8 billion
- Transistor density
- 25,100 K/mm²
- Die size
- 471 mm²
- Package
- BGA-2397
- Released
- 13 September 2016
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
- 1.3 GHz
- Boost clock
- 1.53 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
- 3,840
- Texture mapping units
- 240
- Render output units
- 96
- Streaming multiprocessors
- 30
- L1 cache
- 48 KB
- L2 cache
- 3 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)
- 183.7 GFLOPS
- Single precision (FP32)
- 11.8 TFLOPS
- Double precision (FP64)
- 367.4 GFLOPS
- Pixel rate
- 147 GPixel/s
- Texture rate
- 367 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)
- 250 W
- Suggested power supply
- 600 W
- Power connectors
- 8-pin EPS
- Bus interface
- PCIe 3.0 x16
- Slot width
- Dual-slot
- Dimensions
- 267 mm
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.
- CUDA compute capability
- 6.1
- DirectX
- 12.1
- OpenGL
- 4.6
- Vulkan
- 1.4
- OpenCL
- 3.0
- Shader model
- 6.8
Listings
Where to buy a Tesla P40
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
24 GB
Bandwidth
347 GB/s
Largest model
Mixtral 8x7B
The Tesla P40 carries 24 GB of GDDR5, which covers the mid-sized models most people actually run — about 21.6 GB of it after the runtime and driver reserve their working space.
The memory bus moves 347 GB/s across a 384-bit bus. That is the number that governs generation speed — arithmetic per byte read is small enough that the bus, not the cores, is what everything waits on.
Bandwidth is clock times bus width, and this card clocks its memory at 1.81 GHz. Both halves matter, and neither is visible in a gaming benchmark.
The practical ceiling is Mixtral 8x7B at 46.7B, held at Q3_K_M and running at roughly 26.1 tokens per second.
The chip and how it was built
The Tesla P40 is built on the GP102 graphics processor, using NVIDIA's Pascal architecture, as part of the Tesla Pascal(Pxx) generation.
The chip is manufactured by TSMC, on a 16 nm process, with a die measuring 471 mm², holding 11.8 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 September 2016, roughly 9 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
183.7 GFLOPS
FP64
367.4 GFLOPS
On paper the Tesla P40 reaches 183.7 GFLOPS at half precision and 11.8 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 367.4 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 1.3 GHz at base to 1.53 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 Tesla P40 has 48 KB of L1 cache, backed by 3 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 3,840 shading units, 240 texture mapping units, and 96 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
250 W
The Tesla P40 is rated at 250 W, with a 600 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 267 mm long, and needs 8-pin EPS. 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 Tesla P40
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 Tesla P40
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 Tesla P40
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
Find the model in the table
Every one of the 502 models this Tesla P40 runs is in the table above. Search narrows it by name or by size.
-
02
Decide how long your conversations run
Longer conversations cost memory on top of the weights. With 24 GB to work in, that is frequently the difference between a model fitting and not.
-
03
Choose how far you will compress
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 125 tok/s on Gemma 3 QAT 1B, and which inference software you use moves that by thirty to fifty per cent.
-
05
Check the memory column before committing
Compare what each model needs with the 24 GB this card provides. Tight means it works today; comfortable means it still works when the conversation grows.
-
06
Cross-check against other hardware
Every model name in the table links to its own page, which runs the same calculation across every card we hold. That is where you see whether the Tesla P40 is the right buy for it or merely a card that fits.
Answers
Tesla P40 — common questions
How many tokens per second does a Tesla P40 produce?
It depends on the model. On a Tesla P40 the fastest model we track is Gemma 3 QAT 1B at about 125 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.
Can a Tesla P40 run a 7B model?
Yes. For example a Tesla P40 runs Multi-Token Prediction 7B at Q8_0, using about 7.9 GB of memory and generating around 18.7 tokens per second.
Can a Tesla P40 run a 13B model?
Yes. For example a Tesla P40 runs DeepSeekMoE-16B at Q8_0, using about 17.5 GB of memory and generating around 43.4 tokens per second.
Can a Tesla P40 run a 30B model?
Yes. For example a Tesla P40 runs Nemotron 3-Nano-30B-A3B at Q4_K_M, using about 18.1 GB of memory and generating around 50.7 tokens per second.
How much memory does a Tesla P40 have?
A Tesla P40 has 24 GB of GDDR5 memory. Around a tenth of that is reserved by the inference runtime and the driver, leaving roughly 21.6 GB available for a model and its conversation.
What is the memory bandwidth of a Tesla P40?
The Tesla P40 has 347 GB/s of memory bandwidth, across a 384-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.
What type of memory does a Tesla P40 use?
It uses GDDR5 clocked at 1.81 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.
Who makes the Tesla P40?
The Tesla P40 is a NVIDIA product, with the chip manufactured by TSMC, on a 16 nm process.
When was the Tesla P40 released?
The Tesla P40 was released in September 2016.
How much power does a Tesla P40 use?
The Tesla P40 has a rated board power of 250 W, and a 600 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.
How much cache does a Tesla P40 have?
The Tesla P40 has 48 KB of L1 cache, and 3 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.
What are the TFLOPS of a Tesla P40?
The Tesla P40 is rated at 183.7 GFLOPS at half precision and 11.8 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.
Does the Tesla P40 support CUDA?
Yes. The Tesla P40 reports CUDA compute capability 6.1, which predates tensor cores. Capability 7.0 and above has tensor cores, which modern inference software uses; below that it falls back to slower code paths for quantised models.
What bus interface does the Tesla P40 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.
Is the Tesla P40 good for running local AI models?
Its memory comfortably covers the mid-sized models most people run locally though its bandwidth means generation will feel slow on larger models. In total it runs 502 of the models we track. Whether that is enough depends entirely on which model you want — the table above answers that directly.
Can a Tesla P40 run a model that does not fit in its memory?
Offloading past the card's 24 GB is possible and usually a false economy: the system-memory portion is slow enough to dominate the result.
Would two Tesla P40 cards be twice as fast?
No. A second Tesla P40 doubles the memory to 48 GB, which lets you hold models neither could hold alone, but generation does not split that way. These figures describe one card.
What AI models can a Tesla P40 run?
502 of the 679 open-weight language models we track fit on a Tesla P40 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.
What is the largest AI model a Tesla P40 can run?
The largest model in our catalogue that fits on a Tesla P40 is Mixtral 8x7B at 46.7B parameters, compressed to Q3_K_M. It generates roughly 26.1 tokens per second and needs about 21.0 GB of the card's memory.
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.