CogVideo TPS calculator

Open weights Tsinghua University,Beijing Academy of Artificial Intelligence / BAAI 9.4B parameters May 2022

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

582 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Quadro 6000

6 GB · Q3_K_M · 14.8 tok/s

Fastest card

B200

360 tok/s · 180 GB

Which GPUs can run CogVideo?

Set the inputs, read the answer

Context length tokens

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

Minimum quality

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

Search cards

Card memory

Also show cards that cannot run it

582 cards match

Calculating

					Needs	Quantisation	Fit
360 tok/s 216–577 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	10.8 GB	Q8_0	Comfortable
360 tok/s 216–577 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	10.8 GB	Q8_0	Comfortable
288 tok/s 173–461 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	10.8 GB	Q8_0	Comfortable
288 tok/s 173–461 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	10.8 GB	Q8_0	Comfortable
230 tok/s 138–368 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	10.8 GB	Q8_0	Comfortable
220 tok/s 132–353 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	10.8 GB	Q8_0	Comfortable
220 tok/s 132–353 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	10.8 GB	Q8_0	Comfortable
211 tok/s 127–337 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	10.8 GB	Q8_0	Comfortable
187 tok/s 112–299 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	10.8 GB	Q8_0	Comfortable
187 tok/s 112–299 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	10.8 GB	Q8_0	Comfortable
187 tok/s 112–299 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	10.8 GB	Q8_0	Comfortable
178 tok/s 107–284 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	10.8 GB	Q8_0	Comfortable
155 tok/s 93–248 · low confidence	CMP 170HX 8 GB NVIDIA	8 GB	1,490 GB/s	Sep 2021	6.4 GB	Q4_K_M	Tight
151 tok/s 91–242 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	10.8 GB	Q8_0	Comfortable
151 tok/s 91–242 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	10.8 GB	Q8_0	Comfortable
151 tok/s 91–242 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	10.8 GB	Q8_0	Comfortable
151 tok/s 91–242 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	10.8 GB	Q8_0	Comfortable
151 tok/s 91–242 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	10.8 GB	Q8_0	Comfortable
115 tok/s 69–184 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	10.8 GB	Q8_0	Comfortable
115 tok/s 69–184 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	10.8 GB	Q8_0	Comfortable
102 tok/s 61–163 · low confidence	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.6 GB	Q6_K	Tight
96.1 tok/s 58–154 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	10.8 GB	Q8_0	Comfortable
94.0 tok/s 56–150 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	10.8 GB	Q8_0	Comfortable
91.9 tok/s 55–147 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	10.8 GB	Q8_0	Comfortable
91.9 tok/s 55–147 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	10.8 GB	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

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: Tsinghua University,Beijing Academy of Artificial Intelligence / BAAI
Organisation type: Academia,Academia
Country: China
Published: 29 May 2022
Authors: Wenyi Hong, Ming Ding, Wendi Zheng, Xinghan Liu, Jie Tang

What it does

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

Domain: Video
Task: Video generation, Text-to-video
Base model: CogView2
Numerical format: FP16

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: 9.4B
Training data: 145,000,000,000 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.

How it was established: Operation counting
Fine-tuning compute: 3 × 10¹⁹ FLOP

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)
Training code: Open source

How it is classified

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

Foundation model: Yes
Likely above 10²³ FLOP: Yes
Why it is tracked: Historical significance
Record confidence: Speculative
Citations: 1,058

Sources

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

Reference: CogVideo: Large-scale Pretraining for Text-to-Video Generation via Transformers
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run CogVideo

Ranked by estimated tokens per second, newest card first where speeds tie. Because generation is bound by memory bandwidth, this ordering follows bandwidth rather than any gaming benchmark.

The smallest GPUs that still run CogVideo

The cheapest route in, by memory capacity. A tight fit runs the model but leaves nothing spare for a longer conversation.

What the numbers mean

The hardware side

Minimum card

Quadro 6000

Memory needed

5.3 GB

Fastest

360 tok/s

CogVideo reaches a parameter count of 9.4B. That is small enough that hardware is rarely the obstacle, including on cards several years old. The number of cards we track that can run it: 582.

At the low end it is handled by Quadro 6000, with a memory capacity of 6 GB, running it at a compression of Q3_K_M and producing around 14.8 tokens per second.

At the other end sits B200, generating roughly 360 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Where it came from

CogVideo was published by Tsinghua University,Beijing Academy of Artificial Intelligence / BAAI, in the country recorded as China, during May 2022. The category the publisher falls under is academia,Academia.

It works in the domain of Video, and is recorded as performing the task of video generation, Text-to-video.

Its starting point was an existing base model, CogView2. Most models at this scale are adapted from an existing base rather than built from nothing.

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.

Understanding the speeds

Half the cards that hold it manage more than 22.7 tokens per second. Clearing the ten tokens per second that roughly matches reading speed: 543 of them.

Every weight participates in every token here, so bandwidth is the whole story: the ranking below is effectively a ranking of memory throughput.

Without the attention layout on record, the memory column is an approximation. It is close enough to choose hardware by, and least reliable at long context.

Training and provenance

Training consumed a corpus of around 145,000,000,000 tokens of text.

It is tracked in the underlying dataset for one reason in particular: historical significance.

Step by step

How to choose a GPU for CogVideo

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

01

Start from the memory column

The table lists every card able to hold CogVideo, needing around 5.3 GB at a compression of Q3_K_M. No amount of processing power compensates for a card that cannot hold it.
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 CogVideo.
03

Set a quality floor

The quantisation column varies by card, because a bigger card holds a more accurate copy, reaching a compression of Q3_K_M on the smallest card that fits. 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.
04

Sort by speed

Sort by speed to see how cards rank for CogVideo. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 360 tok/s.
05

Look at the headroom, not just the fit

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

Answers

CogVideo — common questions

CogVideo— can I run it on a GPU holding 12 GB?

Yes. The card GeForce RTX 3080 Ti, holding 12 GB, runs it at a compression of Q8_0, using about 10.8 GB and generating roughly 41.1 tokens per second. The fit is tight.

CogVideo— can I run it on a GPU holding 16 GB?

Yes. The card Tesla V100 SXM2 16 GB, holding 16 GB, runs it at a compression of Q8_0, using about 10.8 GB and generating roughly 50.9 tokens per second. The fit is comfortable.

CogVideo— can I run it on a GPU holding 24 GB?

Yes. The card GeForce RTX 5090 D V2, holding 24 GB, runs it at a compression of Q8_0, using about 10.8 GB and generating roughly 60.4 tokens per second. The fit is comfortable.

CogVideo— 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.

CogVideo— how many parameters does it have?

It has a parameter count of 9.4B. 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.

CogVideo— who created it?

It was published by Tsinghua University,Beijing Academy of Artificial Intelligence / BAAI, based in China, an organisation categorised as academia,Academia.

CogVideo— when was it released?

It was published in May 2022. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

CogVideo— what is it used for?

It works in the domain of Video, and is recorded as handling the task of video generation, Text-to-video. 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.

CogVideo— where can I download it?

The weights are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

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

It can be split between the card and system memory, but it generates painfully slowly that way. The nearest miss we calculate falls short by 1.9 GB. Every figure here assumes the whole model is resident on the card.

CogVideo— would two GPUs run it faster?

A second card roughly doubles the memory available but not the generation rate. The number already able to run it alone: 582. So a second card is rarely the answer here.

CogVideo— why does the quantisation differ between cards?

A larger card holds a more accurate copy. The number of compression levels used across the cards that run it: 4. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

CogVideo— how accurate are these speed estimates?

These are estimates with real error bars, and any of them could reasonably land anywhere in its published range depending on which runtime you use. The fastest result here: 216–577 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

CogVideo— what GPU do I need to run it?

The smallest card in our catalogue that holds it is Quadro 6000, with a memory capacity of 6 GB. It runs the model at a compression of Q3_K_M using about 5.3 GB, and produces roughly 14.8 tokens per second. The number of cards able to run it in total: 582.

CogVideo— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 360 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and the number of cards clearing that: 543.

CogVideo— how much VRAM does it need?

It needs about 5.3 GB at a compression of Q3_K_M, which is what the smallest card that runs it uses. Less compression needs more: the figures in the memory column above are recalculated for each card, because each one holds the least-compressed version it can.

CogVideo— can I run it on a GPU holding 8 GB?

Yes. The card CMP 170HX 8 GB, holding 8 GB, runs it at a compression of Q4_K_M, using about 6.4 GB and generating roughly 155 tokens per second. The fit is tight.

Source

Original publication

Record last updated 25 May 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.

All AI models