MicroNet (Adaptive, Cache) TPS calculator

Open weights Massachusetts Institute of Technology (MIT),Harvard University 8.3M parameters January 2020

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

818 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla C1080

4 GB · Q8_0 · 4,441 tok/s

Fastest card

B200

408,221 tok/s · 180 GB

Which GPUs can run MicroNet (Adaptive, Cache)?

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

818 cards match

Calculating

					Needs	Quantisation	Fit
408,221 tok/s 244,933–653,154 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
408,221 tok/s 244,933–653,154 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
325,975 tok/s 195,585–521,560 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
325,975 tok/s 195,585–521,560 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
260,700 tok/s 156,420–417,120 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
249,525 tok/s 149,715–399,240 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
249,525 tok/s 149,715–399,240 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
238,809 tok/s 143,286–382,095 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
211,943 tok/s 127,166–339,109 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
211,943 tok/s 127,166–339,109 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
211,943 tok/s 127,166–339,109 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
201,049 tok/s 120,629–321,678 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
171,453 tok/s 102,872–274,325 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
171,453 tok/s 102,872–274,325 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
171,453 tok/s 102,872–274,325 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
171,453 tok/s 102,872–274,325 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
171,453 tok/s 102,872–274,325 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
130,549 tok/s 78,329–208,879 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
130,549 tok/s 78,329–208,879 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
108,791 tok/s 65,275–174,065 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
106,469 tok/s 63,882–170,351 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
104,096 tok/s 62,458–166,554 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
104,096 tok/s 62,458–166,554 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
104,096 tok/s 62,458–166,554 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
104,096 tok/s 62,458–166,554 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	0.7 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: Massachusetts Institute of Technology (MIT),Harvard University
Organisation type: Academia,Academia
Country: United States of America
Published: 1 January 2020
Authors: Zhongxia Yan, Hanrui Wang, Demi Guo, Song Han

What it does

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

Domain: Language
Task: Language modeling

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: 8.3M
Training data: tokens

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.

Citations: 8
Benchmark data: MicroNet (Adaptive, Cache)

Sources

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

Reference: MicroNet for Efficient Language Modeling
Last updated: 11 February 2026

The extremes

The ten fastest GPUs that run MicroNet (Adaptive, Cache)

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 MicroNet (Adaptive, Cache)

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

Tesla C1080

Memory needed

0.7 GB

Fastest

408,221 tok/s

MicroNet (Adaptive, Cache) is small enough at 8.3M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

At the low end, a Tesla C1080 handles it — 4 GB, at Q8_0, for about 4,441 tokens per second.

A B200 is the fastest we calculate for it: about 408,221 tokens per second, from 8,000 GB/s of memory bandwidth.

About this model

MicroNet (Adaptive, Cache) was published by Massachusetts Institute of Technology (MIT),Harvard University, in United States of America, in January 2020. The organisation is categorised as academia,Academia.

It works in Language, and is recorded as doing language modeling.

The weights being open is what puts this page in the calculator rather than only in the catalogue: it is a model you can actually hold.

How fast it runs, and why

The median result is around 11,462.9 tokens per second; 818 cards produce text faster than most people read it.

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.

Step by step

How to choose a GPU for MicroNet (Adaptive, Cache)

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

01

Read the memory figure first

The table lists every card that can hold MicroNet (Adaptive, Cache) — around 0.7 GB at Q8_0. That figure, not the card's headline performance, is what decides whether it runs.
02

Match the context to your actual use

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason MicroNet (Adaptive, Cache) stops fitting a card that seemed fine.
03

Choose how far you will compress it

Each card runs the least-compressed copy it can hold — Q8_0 on the smallest card that fits. Setting a floor drops the cards that only manage MicroNet (Adaptive, Cache) by squeezing it further than you would want.
04

Compare tokens per second, not specifications

Sort by speed to see how cards rank for MicroNet (Adaptive, Cache). It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 408,221 tok/s.
05

Check the fit verdict before buying

The fit column separates cards that just manage MicroNet (Adaptive, Cache) from those with room to spare. Buy for the second if the context might grow.
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 MicroNet (Adaptive, Cache) is settled.

Answers

MicroNet (Adaptive, Cache) — common questions

Is MicroNet (Adaptive, Cache) open source?

Its weights are published, so MicroNet (Adaptive, Cache) 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.

How many parameters does MicroNet (Adaptive, Cache) have?

MicroNet (Adaptive, Cache) has 8.3M parameters. 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.

Who created MicroNet (Adaptive, Cache)?

MicroNet (Adaptive, Cache) was published by Massachusetts Institute of Technology (MIT),Harvard University, based in United States of America, categorised as academia,Academia.

When was MicroNet (Adaptive, Cache) released?

MicroNet (Adaptive, Cache) was published in January 2020. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

What is MicroNet (Adaptive, Cache) used for?

MicroNet (Adaptive, Cache) works in Language, and is recorded as handling language modeling. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download MicroNet (Adaptive, Cache)?

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

Can I run MicroNet (Adaptive, Cache) 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 MicroNet (Adaptive, Cache) is rarely worth using. Every figure here assumes the whole model is on the card.

Would two GPUs run MicroNet (Adaptive, Cache) faster?

Capacity adds across cards; throughput does not. Since 818 of the cards we track already hold MicroNet (Adaptive, Cache) on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for MicroNet (Adaptive, Cache)?

A larger card holds a more accurate copy. Across the cards that run MicroNet (Adaptive, Cache), 1 compression levels are used; the floor control above pins it to one.

How accurate are these MicroNet (Adaptive, Cache) speed estimates?

Every figure is derived from memory bandwidth and model size, not benchmarked. That is why each is published as a range such as 244,933–653,154 tok/s on the B200 rather than a single number.

What GPU do I need to run MicroNet (Adaptive, Cache)?

The smallest card in our catalogue that holds MicroNet (Adaptive, Cache) is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.7 GB, and produces roughly 4,441 tokens per second. 818 cards in total can run it.

How fast is MicroNet (Adaptive, Cache) on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 408,221 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 818 of the cards that can run MicroNet (Adaptive, Cache) clear that.

How much VRAM does MicroNet (Adaptive, Cache) need?

About 0.7 GB at Q8_0 compression, 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.

Can I run MicroNet (Adaptive, Cache) on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 0.7 GB and generating roughly 76,031 tokens per second — a comfortable fit.

Can I run MicroNet (Adaptive, Cache) on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 0.7 GB and generating roughly 46,558 tokens per second — a comfortable fit.

Can I run MicroNet (Adaptive, Cache) on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 0.7 GB and generating roughly 57,661 tokens per second — a comfortable fit.

Can I run MicroNet (Adaptive, Cache) on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 0.7 GB and generating roughly 68,377 tokens per second — a comfortable fit.

Source

Original publication

Record last updated 11 February 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