JIUTIAN-139MoE TPS calculator

Open weights China Mobile 38.8B parameters June 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

126 of 818 cards that can run it

Smallest card that fits

Tesla M40 24 GB

24 GB · Q3_K_M · 7.2 tok/s

Fastest card

B200

87.3 tok/s · 180 GB

Which GPUs can run JIUTIAN-139MoE?

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

126 cards match

Calculating

					Needs	Quantisation	Fit
87.3 tok/s 52–140 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	42.2 GB	Q8_0	Comfortable
87.3 tok/s 52–140 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	42.2 GB	Q8_0	Comfortable
69.7 tok/s 42–112 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	42.2 GB	Q8_0	Comfortable
69.7 tok/s 42–112 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	42.2 GB	Q8_0	Comfortable
55.8 tok/s 33–89 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	42.2 GB	Q8_0	Comfortable
53.4 tok/s 32–85 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	42.2 GB	Q8_0	Comfortable
53.4 tok/s 32–85 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	42.2 GB	Q8_0	Comfortable
51.1 tok/s 31–82 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	42.2 GB	Q8_0	Comfortable
45.3 tok/s 27–73 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	42.2 GB	Q8_0	Comfortable
45.3 tok/s 27–73 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	42.2 GB	Q8_0	Comfortable
45.3 tok/s 27–73 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	42.2 GB	Q8_0	Comfortable
43.0 tok/s 26–69 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	42.2 GB	Q8_0	Comfortable
39.5 tok/s 24–63 · low confidence	GeForce RTX 5090 D V2 NVIDIA	24 GB	1,340 GB/s	Aug 2025	19.7 GB	Q3_K_M	Tight
36.7 tok/s 22–59 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	42.2 GB	Q8_0	Comfortable
36.7 tok/s 22–59 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	42.2 GB	Q8_0	Comfortable
36.7 tok/s 22–59 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	42.2 GB	Q8_0	Comfortable
36.7 tok/s 22–59 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	42.2 GB	Q8_0	Comfortable
36.7 tok/s 22–59 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	42.2 GB	Q8_0	Comfortable
36.5 tok/s 22–58 · low confidence	DRIVE A100 PROD NVIDIA	32 GB	1,870 GB/s	May 2020	28.7 GB	Q5_K_M	Tight
36.5 tok/s 22–58 · low confidence	GRID A100A NVIDIA	32 GB	1,870 GB/s	May 2020	28.7 GB	Q5_K_M	Tight
35.9 tok/s 22–58 · low confidence	A30X NVIDIA	24 GB	1,220 GB/s	Apr 2021	19.7 GB	Q3_K_M	Tight
34.9 tok/s 21–56 · low confidence	GeForce RTX 5090 NVIDIA	32 GB	1,790 GB/s	Jan 2025	28.7 GB	Q5_K_M	Tight
34.9 tok/s 21–56 · low confidence	GeForce RTX 5090 D NVIDIA	32 GB	1,790 GB/s	Jan 2025	28.7 GB	Q5_K_M	Tight
29.8 tok/s 18–48 · low confidence	GeForce RTX 3090 Ti NVIDIA	24 GB	1,010 GB/s	Jan 2022	19.7 GB	Q3_K_M	Tight
29.8 tok/s 18–48 · low confidence	GeForce RTX 4090 NVIDIA	24 GB	1,010 GB/s	Sep 2022	19.7 GB	Q3_K_M	Tight

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: China Mobile
Organisation type: Industry
Country: China
Published: 15 June 2024

What it does

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

Domain: Language
Task: Language modeling/generation, Question answering, 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: 38.8B
Training data: 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: 4.4 × 10²³ FLOP
How it was established: Operation counting,Hardware

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 A800 PCIe,Ascend (昇腾) 910B
Chips used: 896
Wall-clock time: 1,464 hours (61 days)

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: Unreleased

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: Confident

Sources

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

Reference: JIUTIAN-139MOE: TECHNICAL REPORT
Last updated: 28 November 2025

The extremes

The ten fastest GPUs for JIUTIAN-139MoE

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 JIUTIAN-139MoE

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 it takes to run this model

Minimum card

Tesla M40 24 GB

Memory needed

19.7 GB

Fastest

87.3 tok/s

With 38.8B parameters, JIUTIAN-139MoE lands in the range a serious desktop card can handle once the weights are compressed. 126 of the cards we track can run it.

The smallest card that holds it is the Tesla M40 24 GB with 24 GB, running it at Q3_K_M and producing around 7.2 tokens per second.

At the other end, a B200 generates roughly 87.3 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

What this model is

JIUTIAN-139MoE was published by China Mobile, in China, in June 2024. It comes out of industry.

It works in Language, and is recorded as doing language modeling/generation, Question answering, Code generation.

Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded.

What decides the speed

Half the cards that hold it manage more than 19.5 tokens per second, and 91 exceed reading speed outright.

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.

How it was trained

Training it took roughly 4.4 × 10²³ FLOP of computation, on NVIDIA A800 PCIe,Ascend (昇腾) 910B — a measure of what producing the model cost, not of how fast it answers.

Step by step

How to choose a GPU for JIUTIAN-139MoE

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

01

Check what it needs before anything else

The table lists every card that can hold JIUTIAN-139MoE — around 19.7 GB at Q3_K_M. 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 JIUTIAN-139MoE 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 — Q3_K_M on the smallest card that fits. Setting a floor drops the cards that only manage JIUTIAN-139MoE by squeezing it further than you would want.
04

Sort by speed

The speed ordering for JIUTIAN-139MoE is effectively an ordering by memory bandwidth, which is why the B200 tops it at 87.3 tok/s.
05

Check the fit verdict before buying

Tight means JIUTIAN-139MoE loads and works, with no room to raise the context later. Comfortable means you can. The difference matters more than a few tokens per second.
06

Open the card you have settled on

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond JIUTIAN-139MoE.

Answers

JIUTIAN-139MoE — common questions

What GPU do I need to run JIUTIAN-139MoE?

The smallest card in our catalogue that holds JIUTIAN-139MoE is the Tesla M40 24 GB, with 24 GB of memory. It runs the model at Q3_K_M using about 19.7 GB, and produces roughly 7.2 tokens per second. 126 cards in total can run it.

How fast is JIUTIAN-139MoE on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 87.3 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 91 of the cards that can run JIUTIAN-139MoE clear that.

How much VRAM does JIUTIAN-139MoE need?

About 19.7 GB at Q3_K_M 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 JIUTIAN-139MoE on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q3_K_M, using about 19.7 GB and generating roughly 39.5 tokens per second — a tight fit.

Is JIUTIAN-139MoE open source?

Its weights are published, so JIUTIAN-139MoE 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 JIUTIAN-139MoE have?

JIUTIAN-139MoE has 38.8B parameters. 38.8B - total parameters 13B - activated 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 JIUTIAN-139MoE?

JIUTIAN-139MoE was published by China Mobile, based in China, categorised as industry.

When was JIUTIAN-139MoE released?

JIUTIAN-139MoE was published in June 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.

What is JIUTIAN-139MoE used for?

JIUTIAN-139MoE works in Language, and is recorded as handling language modeling/generation, Question answering, Code generation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download JIUTIAN-139MoE?

The weights for JIUTIAN-139MoE are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

How much compute was used to train JIUTIAN-139MoE?

Around 4.4 × 10²³ FLOP, on NVIDIA A800 PCIe,Ascend (昇腾) 910B. 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.

Can I run JIUTIAN-139MoE if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part in system memory drags the whole thing down — the nearest miss we calculate is short by 6.2 GB. Our figures for JIUTIAN-139MoE assume it is fully resident.

Would two GPUs run JIUTIAN-139MoE faster?

A second card roughly doubles the memory available but not the generation rate. With 126 cards already able to run JIUTIAN-139MoE alone, the case for pairing is weak.

Why does the quantisation differ between cards for JIUTIAN-139MoE?

Each card is shown running the least-compressed copy it can hold, and JIUTIAN-139MoE appears at 5 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these JIUTIAN-139MoE 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 52–140 tok/s on the B200 rather than a single number.

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