workstation

Nvidia logo Thuê Nvidia RTX PRO 2000 Blackwell.

Blackwell 16GB VRAM 70W 2025
AI models

Models that run on this GPU.

Các mô hình mở trọng số chạy được trên card này để suy luận, với số card ít nhất và độ chính xác cần để tải mỗi mô hình. Tính hằng ngày từ kích thước mô hình và bộ nhớ của card. Hiển thị 60 mô hình lớn nhất trong 119 mô hình chạy được.

?

Dolphin Mistral 24B Venice Edition

by Cognitivecomputations
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

Devstral Medium

by Mistral AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

LFM2-24B-A2B

by Liquid AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

Mistral Small 3.1 24B

by Mistral AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

Mistral Small 3

by Mistral AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

Sarvam M

by Sarvamai
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s
?

Cydonia 24B V4.1

by Thedrummer
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

Mistral Small 3.2 24B

by Mistral AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

Devstral Small 1.1

by Mistral AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

LFM2.5-24B-A2B

by Liquid AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

Magistral Small 2506

by Mistral AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

Devstral Small 2505

by Mistral AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
24B
Throughput
17 tok/s

WizardLM-2 8x22B

by Microsoft
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
22B
Throughput
18 tok/s

gpt-oss-safeguard-20b

by OpenAI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
22B
Throughput
19 tok/s

ERNIE 4.5 21B A3B

by Baidu
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
21B
Throughput
19 tok/s

ERNIE 4.5 21B A3B Thinking

by Baidu
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
21B
Throughput
19 tok/s

Reka Flash 3

by Rekaai
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
21B
Throughput
19 tok/s

GPT-OSS 20B

by OpenAI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
20B (4B active)
Throughput
112 tok/s

InternLM 2.5 20B

by Shanghai AI Lab
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
20B
Throughput
20 tok/s

DeepSeek Coder V2 Lite

by DeepSeek
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
16B (2B active)
Throughput
168 tok/s

Qwen2.5 14B

by Alibaba (Qwen Team)
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
15B
Throughput
27 tok/s

Qwen3 14B Base

by Alibaba (Qwen Team)
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
15B
Throughput
27 tok/s

Cogito V1 Preview Qwen 14B

by Deepcogito
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
15B
Throughput
27 tok/s

Qwen 3 14B

by Alibaba (Qwen Team)
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
15B
Throughput
27 tok/s

Qwen 2.5 14B

by Alibaba (Qwen Team)
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
15B
Throughput
27 tok/s

DeepSeek R1 Distill Qwen 14B

by DeepSeek
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
15B
Throughput
27 tok/s

Phi-4

by Microsoft
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
15B
Throughput
27 tok/s
?

Deepcoder 14B Preview

by Agentica
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
14B
Throughput
29 tok/s

Ministral 3 14B 2512

by Mistral AI
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
14B
Throughput
29 tok/s

Phi-3 Medium

by Microsoft
INT4
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
14B
Throughput
29 tok/s

Code Llama 13B

by Meta AI
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
13B
Throughput
16 tok/s
?

MythoMax 13B

by Gryphe
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
13B
Throughput
16 tok/s

LLaVA 13B

by LLaVA Project
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
13B
Context
4K tokens

Baichuan2-13B

by Baichuan Inc.
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
13B
Throughput
16 tok/s
?

ReMM SLERP 13B

by Undi95
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
13B
Throughput
16 tok/s

Mistral Nemo 12B

by Mistral AI
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
12B
Throughput
17 tok/s

FLUX.1 Dev

by Black Forest Labs
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
12B

FLUX.1 Schnell

by Black Forest Labs
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
12B
?

Rocinante 12B

by Thedrummer
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
12B
Throughput
17 tok/s

Llama Guard 4 12B

by Meta AI
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
12B
Throughput
17 tok/s
?

UnslopNemo 12B

by Thedrummer
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
12B
Throughput
17 tok/s

Gemma 4 12B It

by Google DeepMind
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
12B
Throughput
17 tok/s

Gemma 3 12B

by Google DeepMind
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
12B
Throughput
17 tok/s

Llama 3.2 11B Vision

by Meta AI
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
11B
Context
128K tokens

Qwen3.5-9B

by Alibaba (Qwen Team)
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
10B
Throughput
21 tok/s

Moondream 3.1

by Moondream
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
9B
Throughput
22 tok/s

Moondream 3 Preview

by Moondream
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
9B
Throughput
22 tok/s

Gemma 2 9B

by Google DeepMind
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
9B
Throughput
22 tok/s

Nemotron Nano 9B V2

by Nvidia
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
9B
Throughput
22 tok/s

Ministral 3 8B 2512

by Mistral AI
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
9B
Throughput
23 tok/s

granite-4.2-8b

by IBM Research
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
9B
Throughput
23 tok/s

Granite 4.1 8B

by IBM Research
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
9B
Throughput
23 tok/s

Qwen3 VL 8B Instruct

by Alibaba (Qwen Team)
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
9B
Throughput
23 tok/s

Qwen3 VL 8B Thinking

by Alibaba (Qwen Team)
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
9B
Throughput
23 tok/s

LFM2.5-8B-A1B

by Liquid AI
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
8B
Throughput
24 tok/s

Rnj 1 Instruct

by Essential AI
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
8B
Throughput
24 tok/s
?

UI-TARS 7B

by ByteDance (Seed team)
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
8B
Throughput
24 tok/s

Qwen 3 8B

by Alibaba (Qwen Team)
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
8B
Throughput
25 tok/s

Stable Diffusion 3.5 Large

by Stability AI
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
8B

Hy-MT2-7B

by Tencent
FP8
Required GPUs
1× Nvidia RTX PRO 2000 Blackwell
Tổng VRAM
16 GB
Parameters
8B
Throughput
25 tok/s
Renting for inference?
Pair these models with the cheapest provider on the rental table. See rental rates on the Overview tab →
FAQ

AI models on this GPU.

What AI models can I run on a Nvidia RTX PRO 2000 Blackwell?
The list above shows open-weights models whose memory need fits this card: on one card when the model uses at least a quarter of its 16 GB, or on 2, 4 or 8 of them when it is too large for one. Each row gives the fewest cards and the precision (FP16, FP8 or INT4) that load it.
What's the VRAM minimum to run a model on the Nvidia RTX PRO 2000 Blackwell?
We count the weights (parameters × bytes per weight: FP16 = 2, FP8 = 1, INT4 = 0.5), add the KV cache for an 8K-token context, and add 15% for the runtime. A model fits when that is within 90% of the cards' memory. A 70B model needs about 164 GB at FP16 and about 43 GB at INT4. A longer context needs more.
How does quantization (FP16 vs FP8 vs INT4) affect what fits?
Lower-precision quantization shrinks the memory footprint nearly linearly with the bit count. The trade-off is output quality: FP16 is the reference, FP8 is usually indistinguishable for most prompts, INT8 introduces small quality losses, INT4 is noticeably degraded on reasoning-heavy tasks but fine for chat. The badge on each row tells you which level the recommendation assumes.
Can I fine-tune on the Nvidia RTX PRO 2000 Blackwell or only do inference?
The rows here are for inference. Full fine-tuning needs several times more memory than inference at the same model size, because gradients, optimizer state and activations all live in memory. LoRA and QLoRA cut that a lot. The fine-tune cost tool sizes a run for you.
How are these GPU counts worked out?
We compute them every day from the model's parameter count and the card's memory, with the formula above. They are estimates, not tests we ran. Tokens per second, where shown, is an estimate from the card's memory bandwidth on one card.