Kimi K2 TPS calculator

Open weights Moonshot 1T parameters July 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 cards that can run it

818 cards we hold specifications for

Which GPUs can run Kimi K2?

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
11 July 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, Code generation, Question answering, Quantitative reasoning, Search, Table tasks

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

MoE with 1T total parameters and 32B parameters active per forward pass

Training data
15,500,000,000,000 tokens

"Kimi K2 was pre-trained on 15.5T tokens"

Batch size
67,000,000

"global batch size was held at 67M token"

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
3 × 10²⁴ FLOP

6 FLOP / parameter / token * 32 * 10^9 activated parameters * 15.5 * 10^12 tokens = 2.976e+24 FLOP

How it was established
Operation counting

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
NVIDIA H800 SXM5

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 (separare license is required for entities with 100M+ MAU or $20M+ in annual revenue) for weights and inference code: https://huggingface.co/moonshotai/Kimi-K2-Instruct https://github.com/MoonshotAI/Kimi-K2?tab=readme-ov-file#4-deployment API: https://platform.moonshot.cn/docs/pricing/chat#%E8%AE%A1%E8%B4%B9%E5%9F%BA%E6%9C%AC%E6%A6%82%E5%BF%B5

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
Training cost
Record confidence
Confident

Sources

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

Reference
Kimi K2: Open Agentic Intelligence
Last updated
18 December 2025

What the numbers mean

What it takes to run this model

Kimi K2 reaches a parameter count of 1T. That is beyond what any single graphics card holds. Running it means either splitting it across several cards or renting hardware built for the job, and every card able to hold it alone is a datacentre part. The number that can: 0.

Background

Kimi K2 was published by Moonshot, in the country recorded as China, during July 2025. The category the publisher falls under is industry.

It works in the domain of Language, and is recorded as performing the task of language modeling/generation, Code generation, Question answering, Quantitative reasoning, Search, Table tasks.

Published weights mean the model runs on your machine rather than someone else's, which is what makes the hardware question below answerable at all. On Hugging Face it is published under the organisation moonshotai.

What went into building it

Training it took a computation budget of roughly 3 × 10²⁴ FLOP, on hardware recorded as NVIDIA H800 SXM5. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Training consumed a corpus of around 15,500,000,000,000 tokens of text.

The reason it appears in this catalogue at all: training cost.

Step by step

How to choose a GPU for Kimi K2

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

  1. 01

    Check what it needs before anything else

    The table lists every card able to hold Kimi K2. No amount of processing power compensates for a card that cannot hold it.

  2. 02

    Set the context length you will work at

    Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for Kimi K2.

  3. 03

    Set a quality floor

    The quantisation column varies by card, because a bigger card holds a more accurate copy. Setting a minimum quality drops the cards that only manage it by squeezing further than you would want, and holds the comparison at one level.

  4. 04

    Sort by speed

    Sort by speed to see how cards rank for Kimi K2. It will not match a gaming ordering, because generation is bound by memory bandwidth.

  5. 05

    Read the fit column last

    A tight fit runs, but leaves nothing spare for a longer conversation, in the case of Kimi K2. Comfortable means you can grow the context later. That difference matters more than a few tokens per second, so buy for comfortable if you expect to.

  6. 06

    Check the card from the other side

    Each card page repeats this sweep for every model we hold, answering what else the hardware is good for beyond Kimi K2.

Answers

Kimi K2 — common questions

01

Kimi K2— how accurate are these speed estimates?

Every figure is derived from memory bandwidth and model size, not benchmarked, which is why each is published as a range rather than a single number. One example: the range beneath each figure. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

02

Kimi K2— is it open source?

Its weights are published, so it 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.

03

Kimi K2— how many parameters does it have?

It has a parameter count of 1T. MoE with 1T total parameters and 32B parameters active per forward pass. 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.

04

Kimi K2— who created it?

It was published by Moonshot, based in China, an organisation categorised as industry.

05

Kimi K2— when was it released?

It was published in July 2025.

06

Kimi K2— what is it used for?

It works in the domain of Language, and is recorded as handling the task of language modeling/generation, Code generation, Question answering, Quantitative reasoning, Search, Table tasks. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

07

Kimi K2— where can I download it?

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

08

Kimi K2— how much compute was used to train it?

Training consumed around 3 × 10²⁴ FLOP, on hardware recorded as NVIDIA H800 SXM5. 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.

09

Kimi K2— can I run it if it does not fit in my GPU?

Partly. Layers that do not fit sit in system memory and run at a fraction of the speed, so a mostly-offloaded model is rarely worth using. The nearest miss we calculate falls short by 274.0 GB. Every figure here assumes the whole model is resident on the card.

10

Kimi K2— would two GPUs run it faster?

Capacity adds across cards; throughput does not. The number of cards already holding it on their own: 0. So a second card is rarely the answer here.

11

Kimi K2— why does the quantisation differ between cards?

Each card is shown running the least-compressed copy it can hold. The number of distinct compression levels across the cards that fit it: 1. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

Source

Original publication

Record last updated 18 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.