Solar Open2 250B TPS calculator

Open weights Upstage 250.3B parameters June 2026

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

17 of 818 cards that can run it

Smallest card that fits

Radeon Instinct MI250

128 GB · Q3_K_M · 64.9 tok/s

Fastest card

B200

174 tok/s · 180 GB

Which GPUs can run Solar Open2 250B?

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

17 cards match

Calculating

					Needs	Quantisation	Fit
174 tok/s 104–278 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	135.8 GB	Q4_K_M	Tight
130 tok/s 78–207 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	106.6 GB	Q3_K_M	Tight
113 tok/s 68–181 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	121.2 GB	IQ4_XS	Tight
113 tok/s 68–181 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	121.2 GB	IQ4_XS	Tight
105 tok/s 63–169 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	106.6 GB	Q3_K_M	Tight
75.2 tok/s 45–120 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	252.3 GB	Q8_0	Tight
69.7 tok/s 42–112 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	164.9 GB	Q5_K_M	Tight
69.7 tok/s 42–112 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	164.9 GB	Q5_K_M	Tight
64.9 tok/s 39–104 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	106.6 GB	Q3_K_M	Tight
64.9 tok/s 39–104 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	106.6 GB	Q3_K_M	Tight
63.9 tok/s 38–102 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	194.0 GB	Q6_K	Tight
60.1 tok/s 36–96 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	252.3 GB	Q8_0	Tight
60.1 tok/s 36–96 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	252.3 GB	Q8_0	Tight
54.1 tok/s 32–87 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	106.6 GB	Q3_K_M	Tight
52.9 tok/s 32–85 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	106.6 GB	Q3_K_M	Tight
6.9 tok/s 4–11 · low confidence	GB10 NVIDIA	128 GB	273 GB/s	Oct 2025	106.6 GB	Q3_K_M	Tight
6.9 tok/s 4–11 · low confidence	Jetson T5000 NVIDIA	128 GB	273 GB/s	Aug 2025	106.6 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: Upstage
Organisation type: Industry
Country: Korea (Republic of)
Published: 28 June 2026
Authors: Upstage Team

What it does

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

Domain: Language
Task: Language modeling, Reasoning, Tool use, Translation, Question answering

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: 250.3B
Training data: 11,000,000,000,000 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: 1.1 × 10²⁴ FLOP

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 B200,NVIDIA H200
Chips used: 1,656
Wall-clock time: 1,900 hours (79.2 days)
Power draw: 3.2 MW

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)

How it is classified

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

Why it is tracked: Training cost,Discretionary
Record confidence: Likely

Sources

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

Reference: Solar Open2 250B Hugging Face Model Card
Last updated: 22 July 2026

The extremes

The ten fastest GPUs for Solar Open2 250B

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 Solar Open2 250B

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

Radeon Instinct MI250

Memory needed

106.6 GB

Fastest

174 tok/s

At 250.3B parameters, Solar Open2 250B is beyond what any single graphics card holds. Running it means either splitting it across several cards or renting hardware built for the job — 17 of the cards we track can hold it on their own, and all of them are datacentre parts.

The least hardware that works is a Radeon Instinct MI250. Its 128 GB is enough at Q3_K_M compression, giving roughly 64.9 tokens per second.

A B200 is the fastest we calculate for it: about 174 tokens per second, from 8,000 GB/s of memory bandwidth.

What this model is

Solar Open2 250B was published by Upstage, in Korea (Republic of), in June 2026. The organisation is categorised as industry.

It works in Language, and is recorded as doing language modeling, Reasoning, Tool use, Translation, Question answering.

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached.

What decides the speed

Across every card that can run it, the middle of the range is about 64.9 tokens per second, and 15 of them clear the ten tokens per second that roughly matches reading speed.

Because it routes each token through a subset of its weights, it produces text at the pace of a much smaller model. The catch is memory: all of it still has to fit, so the speed is a bonus rather than a discount on hardware.

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

The training run consumed about 1.1 × 10²⁴ FLOP, on NVIDIA B200,NVIDIA H200. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

Around 11,000,000,000,000 tokens went into training it.

Its inclusion criterion is training cost,Discretionary.

Step by step

How to choose a GPU for Solar Open2 250B

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 Solar Open2 250B — around 106.6 GB at Q3_K_M. That figure, not the card's headline performance, 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 Solar Open2 250B stops fitting a card that seemed fine.
03

Choose how far you will compress it

The quantisation column varies by card, because a bigger card holds a more accurate copy of Solar Open2 250B — Q3_K_M on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Compare tokens per second, not specifications

Ranking by tokens per second for Solar Open2 250B follows memory bandwidth, not core counts, which is why the B200 tops it at 174 tok/s.
05

Look at the headroom, not just the fit

The fit column separates cards that just manage Solar Open2 250B from those with room to spare. Buy for the second if the context might grow.
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 Solar Open2 250B.

Answers

Solar Open2 250B — common questions

How much compute was used to train Solar Open2 250B?

Around 1.1 × 10²⁴ FLOP, on NVIDIA B200,NVIDIA H200. 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 Solar Open2 250B if it does not fit in my GPU?

It can be split between the card and system memory, but Solar Open2 250B generates painfully slowly that way — the nearest miss we calculate is short by 49.4 GB. Nothing on this page assumes offloading.

Would two GPUs run Solar Open2 250B faster?

Capacity adds across cards; throughput does not. Since 17 of the cards we track already hold Solar Open2 250B on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for Solar Open2 250B?

A larger card holds a more accurate copy. Across the cards that run Solar Open2 250B, 6 compression levels are used; the floor control above pins it to one.

How accurate are these Solar Open2 250B speed estimates?

These are estimates with real error bars. The fastest result here, 104–278 tok/s on the B200, could reasonably land anywhere in its published range depending on which runtime you use.

What GPU do I need to run Solar Open2 250B?

The smallest card in our catalogue that holds Solar Open2 250B is the Radeon Instinct MI250, with 128 GB of memory. It runs the model at Q3_K_M using about 106.6 GB, and produces roughly 64.9 tokens per second. 17 cards in total can run it.

How fast is Solar Open2 250B on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 174 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 15 of the cards that can run Solar Open2 250B clear that.

How much VRAM does Solar Open2 250B need?

About 106.6 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.

Is Solar Open2 250B open source?

Its weights are published, so Solar Open2 250B 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 Solar Open2 250B have?

Solar Open2 250B has 250.3B parameters. Total: 250.3B, Active: 14.7B per token. Mixture-of-Experts (320 routed experts + 1 shared expert). 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 Solar Open2 250B?

Solar Open2 250B was published by Upstage, based in Korea (Republic of), categorised as industry.

When was Solar Open2 250B released?

Solar Open2 250B was published in June 2026.

What is Solar Open2 250B used for?

Solar Open2 250B works in Language, and is recorded as handling language modeling, Reasoning, Tool use, Translation, Question answering. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download Solar Open2 250B?

The weights for Solar Open2 250B are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

Source

Original publication

Record last updated 22 July 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