Can the NVIDIA RTX PRO 6000 Blackwell Workstation Edition run Qwen3.8-Flash-Next?

Qwen3.8-Flash-Next · NVIDIA RTX PRO 6000 Blackwell Workstation Edition · 8,192 ctx · whole model residentQ3_K_M sizesize rangeSpeedestimate

IT FITS

Yes. 2 of 9 formats evaluated fit in 96 GB, at 8,192 tokens. No published file exists for this pair, so every size is rebuilt from the pinned architecture. The best quality that fits is Q3_K_M, needing 91.1 GB and running at an estimated 393 tokens per second.

Fitting in memory is not the same as loading. Whether the runtime and version you have supports this architecture and format on this machine has not been tested here.

Find the NVIDIA RTX PRO 6000 Blackwell Workstation Edition: Amazon ↗ · eBay (new and used) ↗Store links may pay us a commission. They never decide which card is suggested — the memory arithmetic does.

2/9formats that fit

96 GBdevice memory

91.1 GBneeded at best quality

~393tokens per second, estimatedFaster than you read

1792GB/s bandwidth

Run it in the cloud

Run Qwen3.8-Flash-Next properly, for cents an hour

The NVIDIA RTX PRO 6000 Blackwell Workstation Edition only holds it at Q3_K_M, at an estimated ~393 tokens per second. A rented card runs a better format at full speed, billed by the second.

Your card Or keep Q4_K_M on your own card: with about 17 GB of it in system RAM, the NVIDIA RTX PRO 6000 Blackwell Workstation Edition runs it at ~95 tokens per second (faster than you read) — free, if you have the RAM. A rented card below holds all of it on the GPU. Size the offload →

Prices · 18:30 UTC, 22 Sept

Recommended · Cheapest128 GB

4× RTX 5090

$1.07per hour

Speed: ~1172 tok/sFaster than you read
Per million tokens: $0.25

Vast.ai · Marketplace · 99.6% reliable

Rent on Vast.ai

Lowest cost per token at 20+ tok/s

Fastest270 GB

1× B300

$7.89per hour

Speed: ~1455 tok/sFaster than you read
Per million tokens: $1.51

RunPod · Secure Cloud

Rent on RunPod

Highest speed estimate

Compare all 11 rentable configurations →

No GPU at all

Use Qwen3.8-Flash-Next by the token

Novita AI$0.47 / M output · $0.15 / M inputOpen on Novita AI
OpenRouter$0.47 / M output · $0.15 / M inputOpen on OpenRouter

Here the rented card is as cheap as the API or cheaper per token ($0.25 per million), and it keeps your data on a machine you control. The API is still the easier start: nothing to set up and nothing to switch off.

First time renting a GPU? How it works, in four steps

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.
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.
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.
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 Qwen3.8-Flash-Next 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 RTX PRO 6000 Blackwell Workstation Edition instead

The models nearest to Qwen3.8-Flash-Next — same lab first, then closest in size — that the NVIDIA RTX PRO 6000 Blackwell Workstation Edition 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.

Qwen3.5-122B-A10BQ4_K_M · 77.9 GB · ~241 tok/s
Qwen3-Coder-NextQ4_K_M · 50.0 GB · ~437 tok/s
Qwen3-Next-80B-A3B-InstructQ4_K_M · 50.0 GB · ~437 tok/s
Qwen3-Next-80B-A3B-ThinkingQ4_K_M · 50.0 GB · ~437 tok/s

How long a conversation. At Q3_K_M, this pair holds 205,825 tokens of context — roughly 154,369 words — needing 96.0 GB. Past that the NVIDIA RTX PRO 6000 Blackwell Workstation Edition runs out of memory, not the model out of context, which would allow 262,144. This is the wall a long chat hits after it has already loaded fine. The nearest round setting below it is 131,072.

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.

Format	Needs	Of 96 GB	Fits	Decode	First token	Basis
FP16	361.3 GB	376%	short by 265.3 GB	does not run	—	size range
Q8_0	192.5 GB	201%	short by 96.5 GB	does not run	—	size range
Q6_K	148.9 GB	155%	short by 52.9 GB	does not run	—	size range
Q5_K_M	129.7 GB	135%	short by 33.7 GB	does not run	—	size range
Q5_0	126.7 GB	132%	short by 30.7 GB	does not run	—	size range
Q4_K_M	111.6 GB	116%	short by 15.6 GB	does not run	—	size range
Q4_0	105.9 GB	110%	short by 9.86 GB	does not run	—	size range
Q3_K_M	91.1 GB	95%	yes	~393 tok/s	no comparable peak	size range
Q2_K	72.4 GB	75%	yes	~472 tok/s	no comparable peak	size range

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.

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 RTX PRO 6000 Blackwell Workstation Edition runs Best models for 96 GB

Where these numbers come from

Qwen3.8-Flash-Next’s architecture is read from its publisher’s own config.json at a pinned revision, and the NVIDIA RTX PRO 6000 Blackwell Workstation Edition’s 96 GB and 1792 GB/s come from the manufacturer’s specification. 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":"qwen-qwen3-8-flash-next","quantization":"Q3_K_M","context":8192,"hardware":"nvidia-rtx-pro-6000-blackwell-workstation"}'

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="qwen-qwen3-8-flash-next" data-quantization="Q3_K_M"
  data-hardware="nvidia-rtx-pro-6000-blackwell-workstation" data-context="8192"></script>

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