Kimi K2 Thinking TPS calculator

Open weights Moonshot 1T parameters November 2025

Each card below is assessed against this model at the context length and minimum quality you choose. Speed is an estimate for a single request, calculated from the card's memory bandwidth and the size of the model once compressed.

Calculated for this model

0 of 818 cards that can run it

Which GPUs can run Kimi K2 Thinking?

Set the inputs, read the answer

A longer conversation needs more memory, which can push this model off smaller cards.

Hides cards that would only fit the model by compressing it below this point.

0 cards match

Calculating
Needs Quantisation Fit

No card in our catalogue can run this model with these settings.

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

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
Moonshot
Organisation type
Industry
Country
China
Published
6 November 2025

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Language
Task
Language modeling/generation, Question answering, Search, Quantitative reasoning, Code generation
Base model
Kimi K2

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
1T

Total Parameters 1T Activated Parameters 32B

Training data
tokens

"Kimi K2 was pre-trained on 15.5T tokens" Training dataset size for Kimi K2 Thinking is not reported

Training compute

The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.

Training compute
4.2 × 10²⁴ FLOP

Assuming the additional post-training contributed between 1% and 100% of Kimi K2's training compute (which we confidently estimate at 2.976e+24), we get a range of 3.0e24 to 6.0e24, and a geometric mean of 4.2e24.

How it was established
Comparison with other models

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Open — downloadable
Model access
Open weights (restricted use)
Training code
Unreleased

modified MIT license (capped by 100M MAU or $20M monthly revenue) https://huggingface.co/moonshotai/Kimi-K2-Thinking

Hugging Face
moonshotai

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Why it is tracked
SOTA improvement

Kimi K2 Thinking sets new records across benchmarks that assess reasoning, coding, and agent capabilities. K2 Thinking achieves 44.9% on HLE with tools, 60.2% on BrowseComp, and 71.3% on SWE-Bench Verified, demonstrating strong generalization as a state-of-the-art thinking agent model.

Record confidence
Likely

Sources

Where this record came from and when it was last checked.

Reference
Introducing Kimi K2 Thinking
Last updated
3 December 2025

What the numbers mean

What it takes to run this model

At 1T parameters, Kimi K2 Thinking is beyond what any single graphics card holds. Running it means either splitting it across several cards or renting hardware built for the job — 0 of the cards we track can hold it on their own, and all of them are datacentre parts.

About this model

Kimi K2 Thinking was published by Moonshot, in China, in November 2025. It comes out of industry.

It works in Language, and is recorded as doing language modeling/generation, Question answering, Search, Quantitative reasoning, Code generation.

It is derived from Kimi K2 rather than trained from scratch, which is the usual way a specialised model is produced.

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached. It is published under the moonshotai organisation on Hugging Face.

How it was trained

Training it took roughly 4.2 × 10²⁴ FLOP of computation — a measure of what producing the model cost, not of how fast it answers.

The reason it appears in this catalogue at all is sOTA improvement.

Step by step

How to choose a GPU for Kimi K2 Thinking

The table above has already assessed every card we hold specifications for against this model. Getting to your answer takes six steps.

  1. 01

    Read the memory figure first

    Every card here has been checked against Kimi K2 Thinking. Capacity is the gate — a card either holds it or it does not.

  2. 02

    Set the context length you will work at

    The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context Kimi K2 Thinking can slip off a card that handles short questions easily.

  3. 03

    Decide how much compression you will accept

    The quantisation column varies by card, because a bigger card holds a more accurate copy of Kimi K2 Thinking. Set a floor to hold the comparison at one level.

  4. 04

    Sort by speed

    Ranking by tokens per second for Kimi K2 Thinking follows memory bandwidth, not core counts.

  5. 05

    Look at the headroom, not just the fit

    A tight fit runs Kimi K2 Thinking but leaves nothing spare for a longer conversation; comfortable has headroom. If you expect to grow the context, buy for comfortable.

  6. 06

    Check the card from the other side

    Following a card through to its own page shows every other model it can hold, which is the question that follows once Kimi K2 Thinking is settled.

Answers

Kimi K2 Thinking — common questions

01

What is Kimi K2 Thinking used for?

Kimi K2 Thinking works in Language, and is recorded as handling language modeling/generation, Question answering, Search, Quantitative reasoning, Code generation. These are the areas it was designed around; they describe intent rather than a hard boundary.

02

Where can I download Kimi K2 Thinking?

Its weights are published under the moonshotai organisation on Hugging Face. We do not host model files — this site calculates what hardware is needed to run them.

03

How much compute was used to train Kimi K2 Thinking?

Around 4.2 × 10²⁴ FLOP. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.

04

Can I run Kimi K2 Thinking if it does not fit in my GPU?

It can be split between the card and system memory, but Kimi K2 Thinking generates painfully slowly that way — the nearest miss we calculate is short by 346.5 GB. Nothing on this page assumes offloading.

05

Would two GPUs run Kimi K2 Thinking faster?

A second card roughly doubles the memory available but not the generation rate. With 0 cards already able to run Kimi K2 Thinking alone, the case for pairing is weak.

06

Why does the quantisation differ between cards for Kimi K2 Thinking?

Because capacity varies, so does how hard Kimi K2 Thinking has to be squeezed — 1 distinct levels appear in the table above. Set a minimum quality to compare at one.

07

How accurate are these Kimi K2 Thinking speed estimates?

These are estimates with real error bars. The fastest result here, the range beneath each figure, could reasonably land anywhere in its published range depending on which runtime you use.

08

Is Kimi K2 Thinking open source?

Its weights are published, so Kimi K2 Thinking can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.

09

How many parameters does Kimi K2 Thinking have?

Kimi K2 Thinking has 1T parameters. Total Parameters 1T Activated Parameters 32B. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

10

Who created Kimi K2 Thinking?

Kimi K2 Thinking was published by Moonshot, based in China, categorised as industry.

11

When was Kimi K2 Thinking released?

Kimi K2 Thinking was published in November 2025.

Source

Original publication

Record last updated 3 December 2025

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.