Inkling TPS calculator

Open weights Thinking Machines 975B parameters July 2026

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 Inkling?

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
Thinking Machines
Organisation type
Industry
Country
United States of America
Published
15 July 2026

What it does

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

Domain
Language, Multimodal, Audio, Video, Vision
Task
Language modeling/generation, Chat

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
975B

975B total, 41B active

Training data
45,000,000,000,000 tokens

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
1.8 × 10²⁴ FLOP

6ND = 41e9 active parameters * 45e12 tokens = 1.845e+24 FLOP

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 GB300 (Blackwell Ultra)

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 (unrestricted)
Hugging Face
thinkingmachines

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
Discretionary
Record confidence
Likely

Sources

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

Reference
Inkling: Our open-weights model
Last updated
18 July 2026

What the numbers mean

What you need to run it

Inkling reaches a parameter count of 975B. 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.

What this model is

Inkling was published by Thinking Machines, in the country recorded as United States of America, during July 2026. It comes out of an organisation categorised as industry.

It works in the domain of Language, Multimodal, Audio, Video, Vision, and is recorded as performing the task of language modeling/generation, Chat.

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

Training and provenance

Producing it required arithmetic totalling around 1.8 × 10²⁴ FLOP, on hardware recorded as NVIDIA GB300 (Blackwell Ultra). That figure measures what producing the model cost, and has no bearing on how fast it answers.

It was trained on a corpus of about 45,000,000,000,000 tokens of text.

Its inclusion criterion: discretionary.

Step by step

How to choose a GPU for Inkling

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

    Start from what it actually needs, which is the requirement of Inkling. That figure, not the headline performance of a card, is what decides whether it runs.

  2. 02

    Decide how long your conversations run

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

  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

    Compare tokens per second, not specifications

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

  5. 05

    Read the fit column last

    The fit column separates cards that just manage it from those with room to spare, in the case of Inkling. 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

    See what else that card runs

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

Answers

Inkling — common questions

01

Inkling— 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.

02

Inkling— why does the quantisation differ between cards?

Because capacity varies, so does how hard it has to be squeezed. The number of distinct levels in the table above: 1. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

03

Inkling— 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.

04

Inkling— 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.

05

Inkling— how many parameters does it have?

It has a parameter count of 975B. 975B total, 41B active. 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.

06

Inkling— who created it?

It was published by Thinking Machines, based in United States of America, an organisation categorised as industry.

07

Inkling— when was it released?

It was published in July 2026.

08

Inkling— what is it used for?

It works in the domain of Language, Multimodal, Audio, Video, Vision, and is recorded as handling the task of language modeling/generation, Chat. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.

09

Inkling— where can I download it?

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

10

Inkling— how much compute was used to train it?

Training consumed around 1.8 × 10²⁴ FLOP, on hardware recorded as NVIDIA GB300 (Blackwell Ultra). 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.

11

Inkling— 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 263.3 GB. Every figure here assumes the whole model is resident on the card.

Source

Original publication

Record last updated 18 July 2026

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.