LLMBOTTLENECK.COM

Can the NVIDIA GeForce RTX 2060 SUPER run dolphin-2.9.1-yi-1.5-34b?

dolphin-2.9.1-yi-1.5-34b · NVIDIA GeForce RTX 2060 SUPER · 8,192 ctx · whole model residentQ2_K sizereconstructed size

IT DOES NOT FIT

No. Even Q2_K, the smallest format evaluated, needs 15.3 GB against 8 GB — short by 7.27 GB. No published file exists for this pair, so every size is rebuilt from the pinned architecture.

0/9formats that fit
8 GBdevice memory
15.3 GBneeded at smallest format
—speed unavailable: does not fit
448GB/s bandwidth
Run it in the cloud

Run dolphin-2.9.1-yi-1.5-34b anyway, on a rented GPU

The NVIDIA GeForce RTX 2060 SUPER cannot hold it, but these rented machines hold the whole model. The ones worth choosing between, priced live and billed by the second.

Your card Keeping about 16 GB of it in system RAM, the NVIDIA GeForce RTX 2060 SUPER runs Q4_K_M at ~3.6 tokens per second (slow) — free, if you have the RAM and the patience. The machines below hold all of it on the GPU. Size the offload →

Prices · 18:30 UTC, 22 Sept
Recommended · Cheapest24 GB

1× RTX 3090

$0.12per hour

Speed
~28 tok/sAbout reading pace
Per million tokens
$1.20

Vast.ai · Marketplace · 99.5% reliable

Rent on Vast.ai

or RunPod $0.50/h

Fastest270 GB

1× B300

$7.89per hour

Speed
~233 tok/sFaster than you read
Per million tokens
$9.39

RunPod · Secure Cloud

Rent on RunPod

Highest speed estimate

Compare all 24 rentable configurations →

First time renting a GPU? How it works, in four steps
  1. Create an account and add credit. Both providers are prepaid: $10 is enough to try any card on this page for hours. A new RunPod account that arrives through a referral link gets a one-time $5 credit when it first adds $10.
  2. Pick the card and the count shown here, and a template: llama.cpp or vLLM for an OpenAI-compatible API, or one with a web chat if you only want to talk to the model. “How to launch it” below gives the exact image and command.
  3. Wait for the download. The weights are fetched on the machine; tens of gigabytes take a few minutes on a data-centre connection. You pay by the second from the moment it starts.
  4. Stop it when you are done. A running machine bills even when idle. A stopped one costs nothing per hour, though its disk is usually still billed until you delete it.
How to launch it

No Q4_K_M file of dolphin-2.9.1-yi-1.5-34b is catalogued here, so there is no exact command to vouch for. Start the machine from the llama.cpp server image ghcr.io/ggml-org/llama.cpp:server-cuda and point -hf at a Q4_K_M build from the publisher or a quantizer you trust — search Hugging Face for one. Check its file size against the memory figure on this page before you rent.

Every machine holds the whole model at Q4_K_M and 8,192 tokens of context, no offload. Speeds are this site’s single-stream decode estimates; prices are what each provider’s own API quoted, on-demand, for the whole machine. Vast hosts below 98% measured reliability are left out.

Referral links Vast.ai, RunPod and Novita pay us a share of what you spend if you sign up through these buttons. It costs you nothing, and it never decides an order or a recommendation: both are computed from the live price and the speed, and options that pay us nothing are listed and recommended on the same terms. How we rank

Runs well on your NVIDIA GeForce RTX 2060 SUPER instead

The models nearest to dolphin-2.9.1-yi-1.5-34b — same lab first, then closest in size — that the NVIDIA GeForce RTX 2060 SUPER holds whole at the standard format and runs at a usable speed. Nearest in size is not the same as equally good; compare them on the task you care about.

Every format evaluated

No published file exists for this pair, so every size is rebuilt from the pinned architecture. The Basis column says which is which for each row. Decode and first-token figures are calibrated estimates, not runs on this card.

FormatNeedsOf 8 GBFitsDecodeFirst tokenBasis
FP1671.6 GB895%short by 63.6 GBdoes not run—reconstructed size
Q8_039.4 GB492%short by 31.4 GBdoes not run—reconstructed size
Q6_K31.0 GB388%short by 23.0 GBdoes not run—reconstructed size
Q5_K_M27.1 GB339%short by 19.1 GBdoes not run—reconstructed size
Q5_026.5 GB332%short by 18.5 GBdoes not run—reconstructed size
Q4_K_M23.5 GB293%short by 15.5 GBdoes not run—reconstructed size
Q4_022.3 GB279%short by 14.3 GBdoes not run—reconstructed size
Q3_K_M19.6 GB245%short by 11.6 GBdoes not run—reconstructed size
Q2_K15.3 GB191%short by 7.27 GBdoes not run—reconstructed size

Sized at 8,192 tokens of context with the whole model resident — weights, the KV cache and the runtime reserve, offload off. Speed is only quoted for a format that fits: a rate for a configuration that cannot load is not a fact about anything.

Why the first-token figure is the same on every row

First token is modelled from the arithmetic the prompt requires, and that count does not change with the weight format — which is why it reads the same on every row. Real prefill does vary by format, because a quantized matmul is a different kernel; this model does not capture that, and the figure should be read as an order of magnitude rather than a ranking between formats.

What the labels mean
published data
Read from a published source — a file's byte count, a model's configuration or a manufacturer's specification — or exact arithmetic on such values. Not a measurement on a machine.
reconstructed size
Weight size reconstructed from the pinned architecture, because no published file exists.
size range
Only a lower and an upper bound are claimed for this weight size.
estimate
Calculated from sourced inputs by a stated method; an estimate, not a measurement.

Ways out

OPTION 1cluster

4× NVIDIA GeForce RTX 2060 SUPER

4 of the same card, tensor parallel, holds it at Q2_K. Often cheaper than one larger card, and it is the option most sizing tools never mention.

Open that configuration →
OPTION 5system RAM

Offload to system memory

Keeping some layers on the host lets it load at all, at a large cost in speed. The calculator sizes the split and prices it.

Size the offload →
What this device does run

The full list is on the NVIDIA GeForce RTX 2060 SUPER page.

Change anything

This page fixes the context at 8,192 tokens and one device. Batch, concurrent users, KV-cache format, clusters and rental cost are all in the calculator, already set to this pairing.

Open in the calculator →Everything the NVIDIA GeForce RTX 2060 SUPER runsBest models for 8 GB

Where these numbers come from

dolphin-2.9.1-yi-1.5-34b’s architecture is read from its publisher’s own config.json at a pinned revision, and the NVIDIA GeForce RTX 2060 SUPER’s 8 GB comes from the specification; its 448 GB/s is a third-party figure, because the manufacturer no longer publishes it, so the speed here is an estimate built on an estimate. How far each figure can be trusted is published, per format and worst case included, on the accuracy scorecard.

Building this into your own product? A free API key returns exactly these figures, and the widget puts this answer on a product page with one script tag.

Use this answer in your own product

As JSON, from the API
curl -s https://llmbottleneck.com/v1/analyze \
  -H "Authorization: Bearer $LLMB_KEY" -H "content-type: application/json" \
  -d '{"model":"dphn-dolphin-2-9-1-yi-1-5-34b","quantization":"Q2_K","context":8192,"hardware":"nvidia-geforce-rtx-2060-super"}'

Same engine, same evidence, every field sourced. Free key in one step →

On your page, as a widget
<script src="https://llmbottleneck.com/widget.js"
  data-model="dphn-dolphin-2-9-1-yi-1-5-34b" data-quantization="Q2_K"
  data-hardware="nvidia-geforce-rtx-2060-super" data-context="8192"></script>

No key needed. Unbranded, with your own buy button, on Pro and Business →

llmbottleneck
catalogue 2026-10-03models 327devices 135