NVIDIA · Blackwell · 16 GB VRAM tier
Best AI Models for NVIDIA RTX PRO 2000 Blackwell (16 GB)
Runs 14B models well; 27–31B barely fits and is slow (13 tok/s).
Entry 70 W low-profile workstation card; 288 GB/s bandwidth is the limiting factor for inference. NVIDIA published only "Q4 2025" availability, so the release date here is the quarter, not a confirmed retail day.
| # | Model | Speed | llmrun Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 29.4 t/s | 29.4 tok/s | Q4_K_M · 6.6 GB | |
| 2 | Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 57.4 t/s | 57.4 tok/s | Q4_K_M · 3.4 GB | |
| 3 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 66.0 t/s | 66.0 tok/s | Q4_K_M · 13.1 GB | |
| 4 | Phi 4Microsoft Q4_K_M · 12.5 GB · 19.7 t/s | 19.7 tok/s | Q4_K_M · 12.5 GB | |
| 5 | Gemma 4 12B ITGoogle Q4_K_M · 9.0 GB · 22.7 t/s | 22.7 tok/s | Q4_K_M · 9.0 GB |
| # | Model | Speed | Coding Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | GPT OSS 20BOpenAIHow to run → Q4_K_M · 13.1 GB · 66.0 t/s | 66.0 tok/s | Q4_K_M · 13.1 GB | |
| 2 | Q4_K_M · 7.9 GB · 35.3 t/s | 35.3 tok/s | Q4_K_M · 7.9 GB | |
| 3 | Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 29.4 t/s | 29.4 tok/s | overall | Q4_K_M · 6.6 GB |
| 4 | Gemma 4 12B ITGoogle Q4_K_M · 9.0 GB · 22.7 t/s | 22.7 tok/s | Q4_K_M · 9.0 GB | |
| 5 | Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 158 t/s | 158 tok/s | Q4_K_M · 5.5 GB |
| # | Model | Speed | Reasoning Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 29.4 t/s | 29.4 tok/s | Q4_K_M · 6.6 GB | |
| 2 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 66.0 t/s | 66.0 tok/s | Q4_K_M · 13.1 GB | |
| 3 | Qwen3 8BAlibaba Q4_K_M · 7.6 GB · 33.9 t/s | 33.9 tok/s | Q4_K_M · 7.6 GB | |
| 4 | Q4_K_M · 7.9 GB · 35.3 t/s | 35.3 tok/s | Q4_K_M · 7.9 GB | |
| 5 | Gemma 4 12B ITGoogle Q4_K_M · 9.0 GB · 22.7 t/s | 22.7 tok/s | Q4_K_M · 9.0 GB |
| # | Model | Speed | Long context Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 29.4 t/s | 29.4 tok/s | Q4_K_M · 6.6 GB | |
| 2 | Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 57.4 t/s | 57.4 tok/s | Q4_K_M · 3.4 GB | |
| 3 | Mistral Nemo Instruct 2407Mistral AI Q4_K_M · 11.0 GB · 23.2 t/s | 23.2 tok/s | Q4_K_M · 11.0 GB | |
| 4 | Q4_K_M · 7.9 GB · 35.3 t/s | 35.3 tok/s | Q4_K_M · 7.9 GB | |
| 5 | Gemma 4 12B ITGoogle Q4_K_M · 9.0 GB · 22.7 t/s | 22.7 tok/s | Q4_K_M · 9.0 GB |
| # | Model | Speed ↓ | llmrun Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | GPT OSS 20BOpenAIHow to run → Q4_K_M · 13.1 GB · 66.0 t/s | 66.0 tok/s | Q4_K_M · 13.1 GB | |
| 2 | Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 57.4 t/s | 57.4 tok/s | Q4_K_M · 3.4 GB | |
| 3 | Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 29.4 t/s | 29.4 tok/s | Q4_K_M · 6.6 GB | |
| 4 | Gemma 4 12B ITGoogle Q4_K_M · 9.0 GB · 22.7 t/s | 22.7 tok/s | Q4_K_M · 9.0 GB | |
| 5 | Phi 4Microsoft Q4_K_M · 12.5 GB · 19.7 t/s | 19.7 tok/s | Q4_K_M · 12.5 GB |
| # | Model | Speed | Vision Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 29.4 t/s | 29.4 tok/s | Q4_K_M · 6.6 GB | |
| 2 | Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 57.4 t/s | 57.4 tok/s | Q4_K_M · 3.4 GB | |
| 3 | Gemma 3 12B ITGoogle Q4_K_M · 10.9 GB · 23.3 t/s | 23.3 tok/s | Q4_K_M · 10.9 GB | |
| 4 | Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 65.9 t/s | 65.9 tok/s | Q4_K_M · 5.0 GB | |
| 5 | Q4_K_M · 6.5 GB · 30.1 t/s | 30.1 tok/s | Q4_K_M · 6.5 GB |
The RTX PRO 2000 Blackwell runs 14B models well. Blackwell generation, released 2025.
RTX PRO 2000 Blackwell strengths and weaknesses
Strengths
- Runs 35B MoE models: Qwen3.6 35B A3B at 75.5 tok/s in 15.7 GB
- Low power draw: 70 W board power; a 150 W PSU is enough
Weaknesses
- 70B models don't fit: Llama 3.3 70B Instruct needs 51.3 GB at Q4 vs 16 GB here
- Slow on large dense models: Qwen3.8 27B at 12.5 tok/s
Output speed depends on memory bandwidth (288 GB/s). Grey rows need more than 16 GB.
- 9BQwen3.5 9B · Q4_K_M · 6.6 GB29.4tok/s
- 14BQwen3 14B · Q4_K_M · 11.9 GB19.7tok/s
- 12.5tok/s
- 75.5tok/s
- 70BLlama 3.3 70B Instruct · not at 4-bitneeds 51.3 GB
- 120B MoEGPT OSS 120B · not at 4-bitneeds 70.8 GB
Estimated speeds: Q4_K_M, 4K context, one user. How we estimate
All 1057 models on the RTX PRO 2000 Blackwell
Featured onlyBest quant that fits, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works
| Model | Quant | VRAM | Speed | Context | Score | Fit |
|---|---|---|---|---|---|---|
DeepSeek R1 Distill Qwen 32B AbliteratedChatReasoning | IQ2_M | 15.7 GB | 15.7 t/s | 131K | D27, Barely runs | |
Deepseek Coder 1.3B BaseChatCode | Q4_K_M | 4.3 GB | 126.5 t/s | 16K | S95, Runs great | |
Mistral Small Instruct 2409ChatFunctions | Q3_K_L | 15.5 GB | 15.4 t/s | 33K | D29, Barely runs | |
| IQ2_M | 14.7 GB | 14.7 t/s | 524K | C36, Usable | ||
CodeLlama 34B Instruct HFChatCode | IQ2_M | 14.9 GB | 15.5 t/s | 16K | C36, Usable | |
Deepseek Coder 6.7B BaseChatCode | Q4_K_M | 12.9 GB | 34.5 t/s | 16K | A71, Runs well | |
Internlm Chat 7BChat | FP16 | 15.4 GB | 12.2 t/s | 2K | D26, Barely runs | |
Medgemma 27B ITVision | Q2_K_S | 15.7 GB | 15.5 t/s | — | D27, Barely runs | |
Qwen2.5 32B InstructChatFunctions | IQ2_M | 15.7 GB | 15.7 t/s | 33K | D27, Barely runs | |
Yi 34BChat | IQ3_S | 15.9 GB | 12.1 t/s | 4K | D19, Barely runs | |
Yi 34B ChatChat | IQ3_S | 15.9 GB | 12.1 t/s | 4K | D19, Barely runs | |
| IQ2_XXS | 15.3 GB | 46.5 t/s | 33K | B53, Decent | ||
Deepseek Coder 33B InstructChatCode | IQ2_M | 15.7 GB | 15.5 t/s | 16K | D27, Barely runs | |
C4ai Command R V01ChatFunctions | IQ2_S | 15.9 GB | 15.6 t/s | — | D25, Barely runs | |
Qwen1.5 32BChat | IQ2_M | 15.6 GB | 15.9 t/s | 33K | C30, Usable |
Specs
- Architecture
- Blackwell
- VRAM
- 16 GB GDDR7
- Memory bandwidth
- 288.0 GB/s
- CUDA cores
- 4,352
- Compute capability
- 12.0
- TDP
- 70 W
- Released
- 2025-10-01
RTX PRO 2000 Blackwell FAQ
What is the largest model the RTX PRO 2000 Blackwell can run?
Gemma 4 26B A4B IT (25.8B) is the largest at 4-bit or better: at Q4_K_S it uses 15.5 GB of 16 GB, and runs at about 64.8 tok/s. At 3-bit, a bigger model fits too: Gemma 2 27B IT (27.2B) at IQ3_M, 15.0 GB.
Can the RTX PRO 2000 Blackwell run Qwen3.5 9B?
Yes. Qwen3.5 9B fits at Q4_K_M, using 6.6 GB of 16 GB, and it is the best model that runs well on this card.
Can the RTX PRO 2000 Blackwell run 70B models?
No. Llama 3.3 70B Instruct needs 51.3 GB at Q4, more than the 16 GB on this card.
How many tokens per second does the RTX PRO 2000 Blackwell give?
About 29.4 tok/s on Qwen3.5 9B at Q4_K_M.
What power supply does the RTX PRO 2000 Blackwell need?
Plan a 150 W PSU (2× the 70 W TDP).
Is 16 GB enough for coding and agents?
Yes for coding: GPT OSS 20B, the best of the 2 coding models that fit, runs at about 66 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.