Ling-Plus ("Bailing") TPS calculator
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
Smallest card that fits
B200
180 GB · IQ4_XS · 28.7 tok/s
Fastest card
B200
28.7 tok/s · 180 GB
Which GPUs can run Ling-Plus ("Bailing")?
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.
7 cards match
Calculating| Needs | Quantisation | Fit | |||||
|---|---|---|---|---|---|---|---|
|
28.7
tok/s
17–46 · low confidence |
B200 NVIDIA | 180 GB | 8,000 GB/s | Jan 2024 | 159.3 GB | IQ4_XS | Tight |
|
17.0
tok/s
10–27 · low confidence |
B300 NVIDIA | 288 GB | 8,000 GB/s | Sep 2025 | 243.7 GB | Q6_K | Tight |
|
14.9
tok/s
9–24 · low confidence |
Radeon Instinct MI300X AMD | 192 GB | 5,325 GB/s | Dec 2023 | 159.3 GB | IQ4_XS | Tight |
|
14.9
tok/s
9–24 · low confidence |
Radeon Instinct MI308X AMD | 192 GB | 5,325 GB/s | Dec 2023 | 159.3 GB | IQ4_XS | Tight |
|
13.6
tok/s
8–22 · low confidence |
Radeon Instinct MI350X AMD | 288 GB | 8,190 GB/s | Jan 2025 | 243.7 GB | Q6_K | Tight |
|
13.6
tok/s
8–22 · low confidence |
Radeon Instinct MI355X AMD | 288 GB | 8,190 GB/s | Jan 2025 | 243.7 GB | Q6_K | Tight |
|
12.2
tok/s
7–20 · low confidence |
Radeon Instinct MI325X AMD | 256 GB | 6,000 GB/s | Oct 2024 | 209.9 GB | Q5_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
- Ant Group
- Organisation type
- Industry
- Country
- China
- Published
- 10 March 2025
- Authors
- Ling Team, Binwei Zeng, Chao Huang, Chao Zhang, Changxin Tian, Cong Chen, Dingnan Jin, Feng Yu, Feng Zhu, Feng Yuan, Fakang Wang, Gangshan Wang, Guangyao Zhai, Haitao Zhang, Huizhong Li, Jun Zhou, Jia Liu, Junpeng Fang, Junjie Ou, Jun Hu, Ji Luo, Ji Zhang, Jian Liu, Jian Sha, Jianxue Qian, Jiewei Wu, Junping Zhao, Jianguo Li, Jubao Feng, Jingchao Di, Junming Xu, Jinghua Yao, Kuan Xu, Kewei Du, Lon…
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
- 290B
- Training data
- 9,000,000,000,000 tokens
290 billion parameters with 28.8 billion activated parameters
"e. To date, we have constructed a high-quality corpus consisting of approximately 9 trillion tokens, distributed across 1 trillion tokens in Chinese, 5.5 trillion in English, and 2.5 trillion in code."
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
- 1.6 × 10²⁴ FLOP
- How it was established
- Operation counting
6 FLOP / parameter / token * 28.8 * 10^9 active parameters * 9 * 10^12 tokens = 1.5552e+24 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
- Unreleased
- Hugging Face
- inclusionAI
MIT license https://huggingface.co/inclusionAI/Ling-plus
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Confident
Sources
Where this record came from and when it was last checked.
- Reference
- Every FLOP Counts: Scaling a 300B Mixture-of-Experts LING LLM without Premium GPUs
- Last updated
- 11 February 2026
The extremes
The ten fastest GPUs for Ling-Plus ("Bailing")
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.
- 01 B200 180 GB · 8,000 GB/s · IQ4_XS 28.7 tok/s
- 02 B300 288 GB · 8,000 GB/s · Q6_K 17.0 tok/s
- 03 Radeon Instinct MI300X 192 GB · 5,325 GB/s · IQ4_XS 14.9 tok/s
- 04 Radeon Instinct MI308X 192 GB · 5,325 GB/s · IQ4_XS 14.9 tok/s
- 05 Radeon Instinct MI350X 288 GB · 8,190 GB/s · Q6_K 13.6 tok/s
- 06 Radeon Instinct MI355X 288 GB · 8,190 GB/s · Q6_K 13.6 tok/s
- 07 Radeon Instinct MI325X 256 GB · 6,000 GB/s · Q5_K_M 12.2 tok/s
The smallest GPUs that still run Ling-Plus ("Bailing")
The cheapest route in, by memory capacity. A tight fit runs the model but leaves nothing spare for a longer conversation.
- 01 B200 180 GB · needs 159.3 GB · IQ4_XS · tight 28.7 tok/s
- 02 Radeon Instinct MI300X 192 GB · needs 159.3 GB · IQ4_XS · tight 14.9 tok/s
- 03 Radeon Instinct MI308X 192 GB · needs 159.3 GB · IQ4_XS · tight 14.9 tok/s
- 04 Radeon Instinct MI325X 256 GB · needs 209.9 GB · Q5_K_M · tight 12.2 tok/s
- 05 B300 288 GB · needs 243.7 GB · Q6_K · tight 17.0 tok/s
- 06 Radeon Instinct MI350X 288 GB · needs 243.7 GB · Q6_K · tight 13.6 tok/s
- 07 Radeon Instinct MI355X 288 GB · needs 243.7 GB · Q6_K · tight 13.6 tok/s
What the numbers mean
What it takes to run this model
Minimum card
B200
Memory needed
159.3 GB
Fastest
28.7 tok/s
At 290B parameters, Ling-Plus ("Bailing") is beyond what any single graphics card holds. Running it means either splitting it across several cards or renting hardware built for the job — 7 of the cards we track can hold it on their own, and all of them are datacentre parts.
The smallest card that holds it is the B200 with 180 GB, running it at IQ4_XS and producing around 28.7 tokens per second.
The quickest result comes from a B200 at around 28.7 tokens per second — its 8,000 GB/s of bandwidth is what buys that.
About this model
Ling-Plus ("Bailing") was published by Ant Group, in China, in March 2025. It comes out of industry.
It works in Language, and is recorded as doing language modeling/generation, Question answering, Code generation.
The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached. It is published under the inclusionAI organisation on Hugging Face.
How fast it runs, and why
Across every card that can run it, the middle of the range is about 14.9 tokens per second, and 7 of them clear the ten tokens per second that roughly matches reading speed.
Being dense, it reads all of itself per token, which is why the ordering by speed below follows the ordering by memory bandwidth so closely.
Its internal architecture is not on file, so memory is approximated from the parameter count and marked accordingly. Expect the real figure to differ, more so at long context.
How it was trained
Training it took roughly 1.6 × 10²⁴ FLOP of computation — a measure of what producing the model cost, not of how fast it answers.
Around 9,000,000,000,000 tokens went into training it.
Step by step
How to choose a GPU for Ling-Plus ("Bailing")
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 Ling-Plus ("Bailing") — around 159.3 GB at IQ4_XS. That figure, not the card's headline performance, is what decides whether it runs.
-
02
Decide how long your conversations run
The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context Ling-Plus ("Bailing") can slip off a card that handles short questions easily.
-
03
Decide how much compression you will accept
Each card runs the least-compressed copy it can hold — IQ4_XS on the smallest card that fits. Setting a floor drops the cards that only manage Ling-Plus ("Bailing") by squeezing it further than you would want.
-
04
Rank by throughput rather than spec sheet
Sort by speed to see how cards rank for Ling-Plus ("Bailing"). It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 28.7 tok/s.
-
05
Read the fit column last
The fit column separates cards that just manage Ling-Plus ("Bailing") from those with room to spare. Buy for the second if the context might grow.
-
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 Ling-Plus ("Bailing") is settled.
Answers
Ling-Plus ("Bailing") — common questions
What is Ling-Plus ("Bailing") used for?
Ling-Plus ("Bailing") works in Language, and is recorded as handling language modeling/generation, Question answering, Code generation. These are the areas it was designed around; they describe intent rather than a hard boundary.
Where can I download Ling-Plus ("Bailing")?
Its weights are published under the inclusionAI organisation on Hugging Face. We do not host model files — this site calculates what hardware is needed to run them.
How much compute was used to train Ling-Plus ("Bailing")?
Around 1.6 × 10²⁴ 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.
Can I run Ling-Plus ("Bailing") if it does not fit in my GPU?
It can be split between the card and system memory, but Ling-Plus ("Bailing") generates painfully slowly that way — the nearest miss we calculate is short by 49.2 GB. Nothing on this page assumes offloading.
Would two GPUs run Ling-Plus ("Bailing") faster?
A second card roughly doubles the memory available but not the generation rate. With 7 cards already able to run Ling-Plus ("Bailing") alone, the case for pairing is weak.
Why does the quantisation differ between cards for Ling-Plus ("Bailing")?
A larger card holds a more accurate copy. Across the cards that run Ling-Plus ("Bailing"), 3 compression levels are used; the floor control above pins it to one.
How accurate are these Ling-Plus ("Bailing") speed estimates?
They are calculated from specifications rather than measured, and each carries a range — 17–46 tok/s on the B200, for instance. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.
What GPU do I need to run Ling-Plus ("Bailing")?
The smallest card in our catalogue that holds Ling-Plus ("Bailing") is the B200, with 180 GB of memory. It runs the model at IQ4_XS using about 159.3 GB, and produces roughly 28.7 tokens per second. 7 cards in total can run it.
How fast is Ling-Plus ("Bailing") on a GPU?
It depends on the card. The quickest we calculate is a B200 at about 28.7 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 7 of the cards that can run Ling-Plus ("Bailing") clear that.
How much VRAM does Ling-Plus ("Bailing") need?
About 159.3 GB at IQ4_XS 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.
Is Ling-Plus ("Bailing") open source?
Its weights are published, so Ling-Plus ("Bailing") 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 Ling-Plus ("Bailing") have?
Ling-Plus ("Bailing") has 290B parameters. 290 billion parameters with 28.8 billion 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 Ling-Plus ("Bailing")?
Ling-Plus ("Bailing") was published by Ant Group, based in China, categorised as industry.
When was Ling-Plus ("Bailing") released?
Ling-Plus ("Bailing") was published in March 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.