Calculate the TPS of the P102-101 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
Ling-lite-1.5 ("Bailing")
16.8B · Q3_K_M · 18.5 tok/s
Fastest model
Gemma 3 QAT 1B
115 tok/s · 1B
Which AI models can run on a P102-101?
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.
381 models match
Calculating| Quantisation | Fit | ||||||
|---|---|---|---|---|---|---|---|
|
115
tok/s
40–231 · low confidence |
Gemma 3 1B | 1B | Mar 2025 | 1.8 GB | 33k tokens | Q8_0 | Comfortable |
|
115
tok/s
40–231 · low confidence |
Gemma 3 QAT 1B | 1B | Apr 2025 | 1.8 GB | 33k tokens | Q8_0 | Comfortable |
|
115
tok/s
40–231 · low confidence |
HGRN 1B (WT 103) ≈ | 1B | Nov 2023 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
115
tok/s
40–231 · low confidence |
LLama 3..2 Typhoon 2 1B ≈ | 1B | Dec 2024 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
115
tok/s
40–231 · low confidence |
OLMo-1B ≈ | 1B | Feb 2024 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
115
tok/s
40–231 · low confidence |
Pythia-1b ≈ | 1B | Apr 2023 | 1.8 GB | 131k tokens ? | Q8_0 | Comfortable |
|
108
tok/s
38–216 · low confidence |
DeepSeekMoE-16B | 16B | Jan 2024 | 8.1 GB | 4k tokens | Q3_K_M | Tight |
|
107
tok/s
37–214 · low confidence |
OpenELM-1.1B ≈ | 1.1B | May 2024 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
105
tok/s
37–210 · low confidence |
DeciCoder-1B ≈ | 1.1B | Aug 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
105
tok/s
37–210 · low confidence |
SantaCoder ≈ | 1.1B | Jan 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
105
tok/s
37–210 · low confidence |
TinyLlama-1.1B (1T token checkpoint) ≈ | 1.1B | Oct 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
105
tok/s
37–210 · low confidence |
TinyLlama-1.1B (3T token checkpoint) ≈ | 1.1B | Oct 2023 | 1.9 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
EXAONE 4.0 (1.2B) ≈ | 1.2B | Jul 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
MinerU2.5 ≈ | 1.2B | Sep 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
Pleias 1.0 1.2B ≈ | 1.2B | Dec 2024 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
96.1
tok/s
34–192 · low confidence |
Pleias-RAG-1B ≈ | 1.2B | Apr 2025 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
93.8
tok/s
33–187 · low confidence |
Llama 3.2 1B | 1.2B | Sep 2024 | 2.2 GB | 131k tokens | Q8_0 | Comfortable |
|
92.4
tok/s
32–185 · low confidence |
MiniCPM-1.2B ≈ | 1.2B | Jun 2024 | 2.0 GB | 131k tokens ? | Q8_0 | Comfortable |
|
88.7
tok/s
31–177 · low confidence |
DeepSeek Coder 1.3B ≈ | 1.3B | Jan 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
88.7
tok/s
31–177 · low confidence |
DeepSeek-VL-1.3B ≈ | 1.3B | Mar 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
88.7
tok/s
31–177 · low confidence |
DigiRL ≈ | 1.3B | Jun 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
88.7
tok/s
31–177 · low confidence |
GLA Transformer 1.3B ≈ | 1.3B | Aug 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
88.7
tok/s
31–177 · low confidence |
Janus 1.3B ≈ | 1.3B | Oct 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
88.7
tok/s
31–177 · low confidence |
Kosmos-2.5 ≈ | 1.3B | Aug 2024 | 2.1 GB | 131k tokens ? | Q8_0 | Comfortable |
|
88.7
tok/s
31–177 · 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
P102-101 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
- 10 GB
- Memory bandwidth
- 320 GB/s
- Memory type
- GDDR5
- Memory bus width
- 320 bit
- Memory clock
- 2 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
- Mining GPUs
- Foundry
- TSMC
- Process size
- 16 nm
- Transistors
- 11.8 billion
- Transistor density
- 25,100 K/mm²
- Die size
- 471 mm²
- Package
- BGA-2397
- Released
- 1 January 2018
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
- 1.56 GHz
- Boost clock
- 1.67 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
- 3,200
- Texture mapping units
- 200
- Render output units
- 80
- Streaming multiprocessors
- 25
- L1 cache
- 48 KB
- L2 cache
- 2.5 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)
- 167 GFLOPS
- Single precision (FP32)
- 10.7 TFLOPS
- Double precision (FP64)
- 334 GFLOPS
- Pixel rate
- 134 GPixel/s
- Texture rate
- 334 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
- 2x 8-pin
- Bus interface
- PCIe 3.0 x4
- 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 P102-101
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
10 GB
Bandwidth
320 GB/s
Largest model
Ling-lite-1.5 ("Bailing")
At 10 GB of GDDR5 the P102-101 is limited to the smaller end of the catalogue. About 9 GB is actually available to a runtime, and a model has to fit entirely inside it before generating anything at all.
The memory bus moves 320 GB/s across a 320-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 2 GHz. Both halves matter, and neither is visible in a gaming benchmark.
The practical ceiling is Ling-lite-1.5 ("Bailing") at 16.8B, held at Q3_K_M and running at roughly 18.5 tokens per second.
The chip and how it was built
The P102-101 is built on the GP102 graphics processor, using NVIDIA's Pascal architecture, as part of the Mining GPUs 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 January 2018, roughly 8 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
167 GFLOPS
FP64
334 GFLOPS
On paper the P102-101 reaches 167 GFLOPS at half precision and 10.7 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 334 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 1.56 GHz at base to 1.67 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 P102-101 has 48 KB of L1 cache, backed by 2.5 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 3,200 shading units, 200 texture mapping units, and 80 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 P102-101 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 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 3.0 x4. 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 P102-101
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 P102-101
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 P102-101
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
The table lists 381 models this P102-101 can run. Search by name, or by size — typing 27b matches on the parameter count even when the name never states it.
-
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 10 GB.
-
03
Choose how far you will compress
By default the table picks the least-compressed copy that fits. Setting a floor removes models that only qualify through heavy compression.
-
04
Look at the range, not just the number
Each speed is an estimate for a single conversation, with a range beneath it — 115 tok/s on Gemma 3 QAT 1B at the top end here. The same card and model vary by thirty to fifty per cent between inference runtimes.
-
05
Read the fit verdict last
A tight fit runs but leaves no room to raise the context later; comfortable has headroom. The memory column shows what each model needs against the 10 GB available.
-
06
Cross-check against other hardware
Each model page repeats this calculation for the whole catalogue. Worth a look before deciding: it shows what else runs the same model, and how the P102-101 compares.
Answers
P102-101 — common questions
Can a P102-101 run a 13B model?
Yes. For example a P102-101 runs DeepSeekMoE-16B at Q3_K_M, using about 8.1 GB of memory and generating around 108 tokens per second.
How much memory does a P102-101 have?
A P102-101 has 10 GB of GDDR5 memory. Around a tenth of that is reserved by the inference runtime and the driver, leaving roughly 9 GB available for a model and its conversation.
What is the memory bandwidth of a P102-101?
The P102-101 has 320 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.
What type of memory does a P102-101 use?
It uses GDDR5 clocked at 2 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 P102-101?
The P102-101 is a NVIDIA product, with the chip manufactured by TSMC, on a 16 nm process.
When was the P102-101 released?
The P102-101 was released in January 2018.
How much power does a P102-101 use?
The P102-101 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 P102-101 have?
The P102-101 has 48 KB of L1 cache, and 2.5 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 P102-101?
The P102-101 is rated at 167 GFLOPS at half precision and 10.7 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 P102-101 support CUDA?
Yes. The P102-101 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 P102-101 use?
It uses PCIe 3.0 x4. 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 P102-101 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 381 of the models we track. Whether that is enough depends entirely on which model you want — the table above answers that directly.
Can a P102-101 run a model that does not fit in its memory?
Only partly. Layers beyond the 10 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.
Would two P102-101 cards be twice as fast?
Pairing P102-101 cards buys headroom rather than pace: 20 GB of combined memory, at roughly the same generation speed as one.
What AI models can a P102-101 run?
381 of the 679 open-weight language models we track fit on a P102-101 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 P102-101 can run?
The largest model in our catalogue that fits on a P102-101 is Ling-lite-1.5 ("Bailing") at 16.8B parameters, compressed to Q3_K_M. It generates roughly 18.5 tokens per second and needs about 8.9 GB of the card's memory.
How many tokens per second does a P102-101 produce?
It depends on the model. On a P102-101 the fastest model we track is Gemma 3 QAT 1B at about 115 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 P102-101 run a 7B model?
Yes. For example a P102-101 runs DeepSeek Coder 6.7B at Q4_K_M, using about 8.3 GB of memory and generating around 39.7 tokens per second.
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.