DeepSeek Coder 6.7B TPS calculator

Open weights DeepSeek,Peking University 6.7B parameters January 2024

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

306 cards that can run it

818 cards we hold specifications for

Smallest card that fits

P102-101

10 GB · Q4_K_M · 39.7 tok/s

Fastest card

B200

506 tok/s · 180 GB

Which GPUs can run DeepSeek Coder 6.7B?

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

306 cards match

Calculating

					Needs	Quantisation	Fit
506 tok/s 430–607	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	11.5 GB	Q8_0	Comfortable
506 tok/s 430–607	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	11.5 GB	Q8_0	Comfortable
404 tok/s 242–646 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	11.5 GB	Q8_0	Comfortable
404 tok/s 242–646 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	11.5 GB	Q8_0	Comfortable
323 tok/s 194–517 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	11.5 GB	Q8_0	Comfortable
309 tok/s 263–371	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	11.5 GB	Q8_0	Comfortable
309 tok/s 263–371	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	11.5 GB	Q8_0	Comfortable
296 tok/s 178–473 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	11.5 GB	Q8_0	Comfortable
263 tok/s 158–420 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	11.5 GB	Q8_0	Comfortable
263 tok/s 158–420 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	11.5 GB	Q8_0	Comfortable
263 tok/s 158–420 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	11.5 GB	Q8_0	Comfortable
249 tok/s 212–299	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	11.5 GB	Q8_0	Comfortable
228 tok/s 194–273	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	8.3 GB	Q4_K_M	Tight
212 tok/s 181–255	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	11.5 GB	Q8_0	Comfortable
212 tok/s 181–255	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	11.5 GB	Q8_0	Comfortable
212 tok/s 181–255	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	11.5 GB	Q8_0	Comfortable
212 tok/s 181–255	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	11.5 GB	Q8_0	Comfortable
212 tok/s 181–255	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	11.5 GB	Q8_0	Comfortable
162 tok/s 97–259 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	11.5 GB	Q8_0	Comfortable
162 tok/s 97–259 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	11.5 GB	Q8_0	Comfortable
135 tok/s 81–216 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	11.5 GB	Q8_0	Comfortable
132 tok/s 79–211 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	11.5 GB	Q8_0	Comfortable
129 tok/s 110–155	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	11.5 GB	Q8_0	Comfortable
129 tok/s 110–155	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	11.5 GB	Q8_0	Comfortable
129 tok/s 110–155	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	11.5 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: DeepSeek,Peking University
Organisation type: Industry,Academia
Country: China
Published: 25 January 2024
Authors: Daya Guo, Qihao Zhu, Dejian Yang, Zhenda Xie, Kai Dong, Wentao Zhang, Guanting Chen, Xiao Bi, Y. Wu, Y.K. Li, Fuli Luo, Yingfei Xiong, Wenfeng Liang

What it does

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

Domain: Language
Task: Code generation

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: 6.7B
Training data: 2,000,000,000,000 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: 8 × 10²² FLOP
How it was established: Operation counting

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
Hugging Face: deepseek-ai

How it is classified

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

Likely above 10²³ FLOP: Yes
Record confidence: Likely

Sources

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

Reference: DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run DeepSeek Coder 6.7B

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 DeepSeek Coder 6.7B

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

What you need to run it

Minimum card

P102-101

Memory needed

8.3 GB

Fastest

506 tok/s

DeepSeek Coder 6.7B reaches a parameter count of 6.7B. 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: 306.

The least hardware that works is P102-101, with a memory capacity of 10 GB, running it at a compression of Q4_K_M and producing around 39.7 tokens per second.

At the other end sits B200, generating roughly 506 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Background

DeepSeek Coder 6.7B was published by DeepSeek,Peking University, in the country recorded as China, during January 2024. It comes out of an organisation categorised as industry,Academia.

It works in the domain of Language, and is recorded as performing the task of code generation.

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 deepseek-ai.

Reading the throughput figures

The median result is around 37.0 tokens per second. Producing text faster than most people read it: 292 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.

Its attention layout is on file, so the memory figures are computed exactly rather than approximated.

What went into building it

Producing it required arithmetic totalling around 8 × 10²² FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Training consumed a corpus of around 2,000,000,000,000 tokens of text.

Step by step

How to choose a GPU for DeepSeek Coder 6.7B

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

Start from what it actually needs, which is the requirement of DeepSeek Coder 6.7B, needing around 8.3 GB at a compression of Q4_K_M. That figure, not the headline performance of a card, is what decides whether it runs.
02

Set the context length you will work at

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason a card that seemed fine stops fitting DeepSeek Coder 6.7B.
03

Set a quality floor

Compression is what makes a model fit smaller cards, at some cost in accuracy, reaching a compression of Q4_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

Rank by throughput rather than spec sheet

Ranking by tokens per second follows memory bandwidth rather than core counts, for DeepSeek Coder 6.7B. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 506 tok/s.
05

Check the fit verdict before buying

A tight fit runs, but leaves nothing spare for a longer conversation, in the case of DeepSeek Coder 6.7B. 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

Open the card you have settled on

Following a card through to its own page shows every other model it can hold, which is the question that follows once you have settled on DeepSeek Coder 6.7B.

Answers

DeepSeek Coder 6.7B — common questions

DeepSeek Coder 6.7B— how much VRAM does it need?

It needs about 8.3 GB at a compression of Q4_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.

DeepSeek Coder 6.7B— 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 Q6_K, using about 9.9 GB and generating roughly 83.8 tokens per second. The fit is tight.

DeepSeek Coder 6.7B— 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 11.5 GB and generating roughly 71.4 tokens per second. The fit is comfortable.

DeepSeek Coder 6.7B— 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 11.5 GB and generating roughly 84.7 tokens per second. The fit is comfortable.

DeepSeek Coder 6.7B— 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.

DeepSeek Coder 6.7B— how many parameters does it have?

It has a parameter count of 6.7B. 6.7B. 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.

DeepSeek Coder 6.7B— who created it?

It was published by DeepSeek,Peking University, based in China, an organisation categorised as industry,Academia.

DeepSeek Coder 6.7B— when was it released?

It was published in January 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

DeepSeek Coder 6.7B— what is it used for?

It works in the domain of Language, and is recorded as handling the task of code generation. 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.

DeepSeek Coder 6.7B— where can I download it?

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

DeepSeek Coder 6.7B— how much compute was used to train it?

Training consumed around 8 × 10²² 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.

DeepSeek Coder 6.7B— 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 1.1 GB. Every figure here assumes the whole model is resident on the card.

DeepSeek Coder 6.7B— would two GPUs run it faster?

Two cards buy memory rather than speed, which matters only if one card cannot hold it. The number that can: 306. So a second card is rarely the answer here.

DeepSeek Coder 6.7B— 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: 3. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

DeepSeek Coder 6.7B— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 430–607 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

DeepSeek Coder 6.7B— what GPU do I need to run it?

The smallest card in our catalogue that holds it is P102-101, with a memory capacity of 10 GB. It runs the model at a compression of Q4_K_M using about 8.3 GB, and produces roughly 39.7 tokens per second. The number of cards able to run it in total: 306.

DeepSeek Coder 6.7B— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 506 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: 292.

Source

Original publication

Record last updated 28 November 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.

All AI models