Prometheus 2 7B on NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip

Can the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip run Prometheus 2 7B?

Yes, it runs well

Runs well. At Q6_K it needs about 19.4 GB of 252 GB usable and makes one decision in about 312 ms.

Best quant: Q6_K
Time per decision: 312 ms
Decisions per second: 3.2/s
Input assumed: 1,024 tokens
Returns: Written feedback, then a score

Best app to run it on the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip

On Windows

LM Studio

Desktop app · runs on llama.cpp, MLX

Point-and-click: search, download and chat in one window.

Open Prometheus 2 7B in LM Studio

llama.cpp Usually the first to support new model architectures.
Ollama Runs the Hugging Face GGUF with ollama run hf.co/…

On Linux

vLLM

Inference server · runs on vLLM

The full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.

vllm serve prometheus-eval/prometheus-7b-v2.0

LM Studio Point-and-click: search, download and chat in one window.
llama.cpp Usually the first to support new model architectures.

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download Prometheus 2 7B

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

huggingface.co/prometheus-eval/prometheus-7b-v2.0-GGUF (opens in a new tab)

ollama run hf.co/prometheus-eval/prometheus-7b-v2.0-GGUF

Original weights · Hugging Face

Full-precision files from Prometheus Eval, for fine-tuning, vLLM, Transformers.

huggingface.co/prometheus-eval/prometheus-7b-v2.0 (opens in a new tab)

No sign-in needed.

Links checked 1 day ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 67Good quality (B) · 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q6_K, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Runs well: How well it runs on the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip: speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

PrecisionEditorial precision tier from the model’s published benchmarks, scaled by the quantization or precision used.	30.3 / 45
Memory fitFull marks when everything sits in GPU memory; fewer when part spills to system RAM.	25.0 / 25
SpeedTime for one decision against a 100 ms target: reading the input plus any tokens the model writes.	8.5 / 15
ClarityHow easy the answer is to use in code: a probability or score earns the most, written feedback before a score the least.	9.0 / 15
Match score	72.8 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every quantization on the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q4_K_M`	17.9 GB	252 GB	257 ms	Runs well	64Good quality (B) · 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	18.6 GB	252 GB	284 ms	Runs well	65Good quality (B) · 65 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q6_K`	19.4 GB	252 GB	312 ms	Runs well	67Good quality (B) · 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q8_0`	21.1 GB	252 GB	377 ms	Runs well	68Good quality (B) · 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`F16`	27.9 GB	252 GB	664 ms	Slow	68Good quality (B) · 68 of 100: how capable the model is at F16, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)

Times are for one request with a 1,024-token input and about 150 tokens written. Batching servers process many requests at once.

Other hardware that runs it

64Good quality (B) · 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA B200 Q4_K_M · 179 ms · Runs well
67Good quality (B) · 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. AMD Instinct MI350X Q6_K · 276 ms · Runs well
67Good quality (B) · 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. AMD Instinct MI325X Q6_K · 324 ms · Runs well
67Good quality (B) · 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA H200 141GB SXM Q6_K · 347 ms · Runs well
67Good quality (B) · 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA H200 NVL Q6_K · 355 ms · Runs well
67Good quality (B) · 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. AMD Instinct MI300X Q6_K · 357 ms · Runs well

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.