VOOZH about

URL: https://willitrunai.com/gpus/rtx-pro-6000-blackwell-server-96gb

โ‡ฑ AI Models for RTX PRO 6000 Blackwell Server Edition 96GB โ€” What Runs on 96GB VRAM


NVIDIA

RTX PRO 6000 Blackwell Server Edition 96GB

RTX PRO BlackwellDatacenterBlackwellPCIe 5CUDA

Operating mode

Choose the operating mode for this hardware

Use this to bias workload recommendations toward responsiveness, background autonomy, lighter serving, or multi-GPU scale-out.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

See Full AI Tier List for RTX PRO 6000 Blackwell Server Edition 96GB โ†’

About this GPU for AI

The RTX PRO 6000 Blackwell Server Edition is NVIDIA's most capable workstation-class accelerator, packing 96 GB of GDDR7 VRAM on Blackwell architecture for professional AI and visualization workloads. Its 120 TFLOPS of FP16 compute and 1,597 GB/s bandwidth make it suited for running 70B parameter models at Q4 with headroom to spare. As a PCIe 5.0 card, it slots into standard server platforms without the infrastructure requirements of SXM or NVLink systems. It bridges the gap between consumer workstations and full datacenter deployments.

Beyond LLMs

AI Capability Matrix

What AI tasks this GPU can handle โ€” from text generation to image and video creation.

CapabilityStatusRepresentative ModelDetail
LLM Chat (7B)Runs nativelyLlama 3.1 8B Q4โ€”
LLM Coding (30B)Runs nativelyQwen 3 30B Q4โ€”
LLM Large (70B)
massive-vramblackwell-architecturepcie-form-factorprofessional-grade

Specifications

Compute
FP16120 TFLOPS
INT84000 TOPS
ArchitectureBlackwell
Memory
VRAM96 GB
Bandwidth1597 GB/s
General
FamilyRTX PRO Blackwell
SegmentDatacenter
InterconnectPCIe 5
Compute PlatformCUDA
MSRP$9,999

Key Features

96 GB GDDR7 VRAM โ€” largest capacity in the RTX PRO lineupBlackwell architecture with 4th-gen Tensor Cores (FP4/FP8/FP16)1,597 GB/s memory bandwidth120 TFLOPS FP16 / 4,000 INT8 TOPSPCIe 5.0 x16 โ€” drop-in for modern server platforms300W TDP class โ€” single-slot power budget

For AI Workloads

Strengths
  • 96 GB VRAM fits 70B models at Q4 and 34B models at FP16 on a single card
  • Blackwell Tensor Cores with FP4 support deliver strong inference throughput per watt
  • Standard PCIe 5.0 form factor works in any modern server โ€” no proprietary baseboard needed
  • Commercially available at a fraction of H100/A100 pricing
Considerations
  • GDDR7 bandwidth (1,597 GB/s) significantly below HBM-based datacenter GPUs like A100/H100
  • No NVLink support limits multi-GPU scaling to PCIe peer-to-peer speeds
  • Not suited for large-scale distributed training across GPU clusters
  • ~$10K price point still steep for individual researchers or small teams

Architecture

Blackwell

Blackwell is NVIDIA's fifth-generation RTX architecture, built on TSMC's 4NP process. It introduces 5th-generation Tensor Cores with native FP4 precision support, enabling double the inference throughput per watt compared to Ada Lovelace's FP8 operations. Key innovations include the Neural Rendering Pipeline for AI-driven shading and the debut of GDDR7 memory in consumer GPUs.

AI Relevance

FP4 Tensor Cores deliver the highest tokens-per-watt efficiency in any consumer architecture. Native FP4 quantization means models can run at lower precision with minimal quality loss, effectively doubling the effective VRAM for model weights.

Process: TSMC 4NPPlatform: CUDATensor Cores: Gen 5Precisions: FP32, FP16, BF16, FP8, FP4, INT8, INT4

Recommendations by Workload

Chat

S

Qwen 3.5 27B

Qwen 3.5 27B matches Chat and keeps a practical fit profile. It is a recent-generation family, which helps on current local SOTA workloads. It fits natively with comfortable headroom. Context coverage stays within the requested workload envelope. Known distribution channels: huggingface, ollama, lm-studio.

Decode 55.2 tok/s ยท 131K ctx ยท llama.cppEST.
41.0 GB / 96.0 GB VRAM

Coding

S

Qwen3-Coder-Next

Qwen3-Coder-Next is a specialized fit for Coding. It is a recent-generation family, which helps on current local SOTA workloads. It fits natively with comfortable headroom. Context coverage stays within the requested workload envelope. Known distribution channels: huggingface, ollama, lm-studio.

Decode 70.8 tok/s ยท 217K ctx ยท llama.cppEST.
77.6 GB / 96.0 GB VRAM

Agentic Coding

Full Model Compatibility

๐Ÿ‘ Alibaba
Qwen3-Coder-Next
S95
80B60.8 GB91 tok/s256K ctx
moe
๐Ÿ‘ Alibaba
Qwen 3.5 122B A10B
S95
122B87.4 GB54 tok/s73K ctx
moe
๐Ÿ‘ Mistral
Mistral Small 4 119B
S

Just out of reach

Models you could run with an upgrade

High-quality models that need a bit more memory

Image & Video Generation

Diffusion Model Compatibility

51 of 52 models can generate images or video on your RTX PRO 6000 Blackwell Server Edition 96GB

ModelMax ResolutionGen TimeGrade
SD TurboImage512ร—512300msS
Stable Diffusion 1.5Image512ร—768600msS
Realistic Vision v5.1Image512ร—768600msS
DreamShaper 8Image512ร—768600msS
LCM DreamShaper v7

Upgrade paths

Upgrade from RTX PRO 6000 Blackwell Server Edition 96GB

See what you unlock with more powerful hardware

Upgrade options

Upgrade options

๐Ÿ‘ NVIDIA
NVIDIA H200 141GBNext step up
141 GB VRAM (+45)4800 GB/s (+3203)
B
Unlocks 2 additional models that do not fit on the current setup.Unlocks Qwen 3 235B A22B, MiniMax M2.7+55% faster avg

Unlocks 2 additional models that do not fit on the current setup.

Lifts average decode speed across fitting models by about 55%.

~$30,000 MSRP

๐Ÿ‘ NVIDIA
NVIDIA B200 180GBNVIDIA upgrade
180 GB VRAM (+84)8000 GB/s (+6403)
B
Unlocks 8 additional models that do not fit on the current setup.Unlocks DeepSeek V4 Flash, Qwen 3 235B A22B, MiniMax M2.7+5 more ยท +89% faster avg

Unlocks 8 additional models that do not fit on the current setup.

Lifts average decode speed across fitting models by about 89%.

~$30,000 MSRP

AMD Instinct MI325X 256GBBiggest leap
256 GB VRAM (+160)6000 GB/s (+4403)
B
Unlocks 12 additional models that do not fit on the current setup.Unlocks Qwen 3.5 397B A17B, DeepSeek V4 Flash, Qwen 3 235B A22B+9 more ยท +58% faster avg

Unlocks 12 additional models that do not fit on the current setup.

Lifts average decode speed across fitting models by about 58%.

~$20,000 MSRP

AMD Instinct MI350X 288GBBest value
288 GB VRAM (+192)8000 GB/s (+6403)
B
Unlocks 13 additional models that do not fit on the current setup.Unlocks Qwen 3.5 397B A17B, DeepSeek V4 Flash, Qwen 3 235B A22B+10 more ยท +77% faster avg

Unlocks 13 additional models that do not fit on the current setup.

Lifts average decode speed across fitting models by about 77%.

~$8,000 MSRP

Frequently Asked Questions

Compare with similar

RTX PRO 6000 Blackwell Server Edition 96GB vs RTX PRO 6000 Blackwell Workstation Edition 96GBRTX PRO 6000 Blackwell Server Edition 96GB vs NVIDIA GH200 96GBRTX PRO 6000 Blackwell Server Edition 96GB vs NVIDIA H20 96GB
Compare this GPUCompare with another GPU โ†’
96GB
VRAM
1.6kGB/s
Bandwidth
120TFLOPS
FP16 Compute
4kTOPS
INT8 Inference
$9,999 MSRP
RTX PRO 6000 Blackwell Server Edition 96GBCategory AvgNVIDIA H200 141GB
Runs natively
Llama 3.1 70B Q4
โ€”
Image Gen (SDXL)Runs nativelySDXL 1.0 FP16~~2.5s per image
Image Gen (Flux)Runs nativelyFlux.1 Dev FP16~~11.1s per image
Image Gen (SD 3.5)Runs nativelySD 3.5 Large FP16~~13.6s per image
Video Short (25f)Runs nativelyLTX Video 2B~~2.1s/frame
Video Long (100f)Tight fitWan Video 14B~~6.3s/frame
S

Qwen3-Coder-Next

Qwen3-Coder-Next is a specialized fit for Agentic Coding. It is a recent-generation family, which helps on current local SOTA workloads. It fits natively with comfortable headroom. Context coverage stays within the requested workload envelope. Known distribution channels: huggingface, ollama, lm-studio.

Decode 90.6 tok/s ยท 256K ctx ยท llama.cppEST.
62.2 GB / 96.0 GB VRAM

Reasoning

S

Qwen3-Coder-Next

Qwen3-Coder-Next matches Reasoning and keeps a practical fit profile. It is a recent-generation family, which helps on current local SOTA workloads. It fits natively with comfortable headroom. Context coverage stays within the requested workload envelope. Known distribution channels: huggingface, ollama, lm-studio.

Decode 70.8 tok/s ยท 217K ctx ยท llama.cppEST.
77.6 GB / 96.0 GB VRAM

RAG

S

Qwen 3.5 27B

Qwen 3.5 27B matches RAG and keeps a practical fit profile. It is a recent-generation family, which helps on current local SOTA workloads. It fits natively with comfortable headroom. Context coverage stays within the requested workload envelope. Known distribution channels: huggingface, ollama, lm-studio.

Decode 55.2 tok/s ยท 131K ctx ยท llama.cppEST.
45.7 GB / 96.0 GB VRAM
93
119B88.5 GB59 tok/s38K ctx
moe
๐Ÿ‘ Alibaba
Qwen 2.5 VL 72B
S92
72B59.3 GB33 tok/s33K ctx
dense
๐Ÿ‘ Alibaba
Qwen 3.6 35B A3B
S92
35B36.0 GB171 tok/s250K ctx
+1moe
๐Ÿ‘ Mistral
Devstral 2 123B Instruct
S92
123B90.9 GB19 tok/s31K ctx
dense
๐Ÿ‘ Alibaba
Qwen3-Coder 30B A3B Instruct
S92
30.5B30.6 GB203 tok/s256K ctx
moe
๐Ÿ‘ Alibaba
Qwen 3.5 27B
S91
27B30.1 GB88 tok/s131K ctx
dense
๐Ÿ‘ Alibaba
Qwen3-VL 30B A3B Instruct
S91
30B30.3 GB210 tok/s256K ctx
moe
๐Ÿ‘ Alibaba
Qwen 3.5 35B A3B
S91
35B33.3 GB185 tok/s131K ctx
moe
๐Ÿ‘ Alibaba
Qwen 3 32B
S90
32B33.9 GB75 tok/s131K ctx
dense
๐Ÿ‘ Cohere
Command A 111B
S90
111B82.1 GB22 tok/s73K ctx
dense
๐Ÿ‘ Mistral
Magistral Small 2507
S89
24B27.6 GB99 tok/s131K ctx
dense
๐Ÿ‘ OpenAI
GPT-OSS 120B
S89
117B86.8 GB20 tok/s46K ctx
dense
๐Ÿ‘ Mistral
Devstral Small 2 24B Instruct
S89
24B27.6 GB99 tok/s256K ctx
dense
๐Ÿ‘ Alibaba
Qwen 3 30B A3B
S89
30.5B30.6 GB203 tok/s131K ctx
moe
๐Ÿ‘ Alibaba
Qwen 3.6 27B
S89
27B27.9 GB55 tok/s262K ctx
+1dense
๐Ÿ‘ NVIDIA
Nemotron 3 Nano 30B
S88
30B31.2 GB79 tok/s131K ctx
dense
๐Ÿ‘ Mistral AI
Pixtral Large 124B
S88
124B91.5 GB19 tok/s29K ctx
dense
๐Ÿ‘ Mistral
Leanstral 119B A6B
S88
119B91.9 GB54 tok/s24K ctx
moe
๐Ÿ‘ Mistral
Devstral Small 1.1
S88
24B27.6 GB99 tok/s131K ctx
dense
๐Ÿ‘ Alibaba
Qwen 3.5 9B
S88
9B18.2 GB126 tok/s131K ctx
dense
๐Ÿ‘ Alibaba
Qwen 3 14B
S87
14B21.5 GB170 tok/s131K ctx
dense
๐Ÿ‘ NVIDIA
Nemotron Cascade 2 30B A3B
S87
30B31.7 GB207 tok/s262K ctx
moe
๐Ÿ‘ Microsoft
Phi-4-reasoning-plus 14B
S87
14.7B22.5 GB161 tok/s33K ctx
dense
๐Ÿ‘ Google
Gemma 4 31B
S86
30.7B43.9 GB47 tok/s73K ctx
dense
๐Ÿ‘ OpenAI
GPT-OSS 20B
S86
21B25.8 GB258 tok/s128K ctx
moe
๐Ÿ‘ Alibaba
Qwen 3 8B
S86
8B17.6 GB112 tok/s131K ctx
dense
๐Ÿ‘ LG AI
EXAONE 4.0 32B
A84
32B33.9 GB74 tok/s131K ctx
dense
๐Ÿ‘ Alibaba
Qwen 3.5 4B
A84
4B15.1 GB56 tok/s131K ctx
dense
๐Ÿ‘ Google
Gemma 4 26B A4B
A83
25.2B29.5 GB218 tok/s256K ctx
moe
๐Ÿ‘ Mistral
Ministral 3 14B
A82
14B21.5 GB169 tok/s262K ctx
multimodal
๐Ÿ‘ NVIDIA
Nemotron Nano 8B
A81
8B17.3 GB112 tok/s131K ctx
dense
๐Ÿ‘ Microsoft
Phi-4 Mini Reasoning 4B
A80
3.8B14.3 GB53 tok/s131K ctx
dense
๐Ÿ‘ Jina AI
Jina Embeddings v3
A74
0.57B13.6 GB8 tok/s8K ctx
dense
๐Ÿ‘ BAAI
BGE M3
A73
0.57B12.8 GB8 tok/s8K ctx
dense
๐Ÿ‘ Alibaba
Qwen 3.5 397B A17B
F0
397B255.5 GB3 tok/s4K ctx
moe
๐Ÿ‘ Moonshot AI
Kimi K2.5
F0
1000B627.9 GB2 tok/s4K ctx
moe
๐Ÿ‘ Moonshot AI
Kimi K2.6
F0
1000B627.9 GB2 tok/s4K ctx
+1moe
๐Ÿ‘ DeepSeek
DeepSeek V4 Pro
F0
1600B874.4 GB2 tok/s4K ctx
moe
๐Ÿ‘ DeepSeek
DeepSeek V4 Flash
F0
284B169.8 GB7 tok/s4K ctx
moe
๐Ÿ‘ Z.ai
GLM-5.1
F0
754B489.5 GB2 tok/s4K ctx
moe
๐Ÿ‘ Z.ai
GLM-5
F0
744B483.4 GB2 tok/s4K ctx
moe
๐Ÿ‘ DeepSeek
DeepSeek V3.2
F0
671B420.3 GB2 tok/s4K ctx
moe
๐Ÿ‘ Alibaba
Qwen 3 235B A22B
F0
235B156.7 GB8 tok/s4K ctx
moe
๐Ÿ‘ Alibaba
Qwen3-Coder 480B A35B Instruct
F0
480B306.2 GB2 tok/s4K ctx
moe
MiniMax M2.7
F0
230B154.6 GB9 tok/s4K ctx
moe
๐Ÿ‘ DeepSeek
DeepSeek Coder V2 236B
F0
236B213.1 GB4 tok/s4K ctx
moe
๐Ÿ‘ DeepSeek
DeepSeek R1 671B
F0
671B479.4 GB2 tok/s4K ctx
moe
๐Ÿ‘ DeepSeek
DeepSeek V3.1 671B
F0
671B479.4 GB2 tok/s4K ctx
moe
Image
512ร—768
200ms
S
PixArt-SigmaImage1024ร—1024~2.5sS
FramePack I2VVideo1280ร—720~4.5s/frameS
SDXL TurboImage512ร—512300msS
SDXL LightningImage1024ร—1024900msS
Stable Diffusion XL 1.0Image1024ร—1024~2.5sS
Playground v2.5Image1024ร—1024~3.7sS
RealVisXL v5.0Image1024ร—1024~2.8sS
DreamShaper XLImage1024ร—1024~2.8sS
Juggernaut XL v9Image1024ร—1024~2.8sS
Animagine XL 3.1Image1024ร—1024~2.8sS
Pony Diffusion V6 XLImage1024ร—1024~2.8sS
Animagine XL 4.0Image1024ร—1024~2.8sS
Illustrious XLImage1024ร—1024~2.8sS
Wan Video 2.1 1.3BVideo480ร—832~1.8s/frameS
Stable Diffusion 3.5 MediumImage1024ร—1024~4.3sS
Flux.2 Klein 4BImage1024ร—1024700msS
LTX Video 2BVideo1280ร—720~2.1s/frameS
KolorsImage1024ร—1024~4.9sS
Stable CascadeImage1024ร—1024~6.2sS
AuraFlow v0.3Image1536ร—1536~11.1sS
Stable Diffusion 3.5 LargeImage1024ร—1024~13.6sS
Stable Diffusion 3.5 Large TurboImage1024ร—1024~2.5sS
CogVideoX 2BVideo720ร—480~2.1s/frameS
HunyuanVideoVideo720ร—1280~4.5s/frameS
ChromaImage1024ร—1024~2.5sS
Z-Image TurboImage1536ร—1536~2.6sS
Flux.1 DevImage1024ร—1024~11.1sS
Flux.1 SchnellImage1024ร—1024~2.2sS
LTX Video 13BVideo1280ร—720~4.5s/frameS
Flux.1 Kontext DevImage1024ร—1024~12.4sS
AnimateDiff v1.5.3Video512ร—768~1.1s/frameS
Cosmos Diffusion 7BVideo1024ร—576~3.5s/frameS
CogVideoX 5BVideo720ร—480~3.1s/frameS
Wan2.2 TI2V 5BVideo832ร—480~3.1s/frameS
Flux.2 Klein 9BImage1024ร—1024~1.2sS
Flux.1 Fill DevImage1024ร—1024~10.5sS
Mochi 1 PreviewVideo848ร—480~4.1s/frameS
HunyuanVideo 1.5Video720ร—1280~3.8s/frameS
Helios 14BVideo1280ร—720~4.7s/frameS
SkyReels V2 14BVideo1280ร—720~4.7s/frameS
Wan Video 2.1 14BVideo720ร—1280~4.7s/frameS
Wan Video 2.2 14BVideo720ร—1280~4.7s/frameS
Qwen ImageImage1024ร—1024~4.2sS
Qwen Image EditImage1024ร—1024~4.2sS
Flux.2 DevImage1024ร—1024~1m 57sS
MAGI-1Video1280ร—720~5.8s/frameS
HunyuanImage 3.0Image256ร—256~7.3sF

Image models estimated at 1024ร—1024 (28 steps, FP16). Video models estimated at 768ร—512 (25 frames, 30 steps, FP16). Actual performance varies with runtime and system load.

Buying advice

Should you buy RTX PRO 6000 Blackwell Server Edition 96GB for local AI?

Excellent choice for local AI

Runs 36 of 50 top models well โ€” a strong all-rounder for local inference.

96.0 GB

VRAM

$9,999

MSRP

$104/GB

Cost per GB VRAM

Best models for this GPU

What will limit you first

This setup is broadly balanced for this model.

No major red flags

This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.

Best upgrade itinerary

Unlocks 2 additional models that do not fit on the current setup.

Want more headroom? NVIDIA H200 141GB (141.0 GB VRAM) is the next step up.