readygpu

NVIDIA RTX 6000 Ada Generation vs NVIDIA RTX PRO 5000 Blackwell 48GB

For running models locally, priced in Canada. At 128k context the first runs 60 models fully in memory, the second 60.

In short

Both have 48 GB of memory for models; the NVIDIA RTX 6000 Ada Generation has 960 GB/s of bandwidth and the NVIDIA RTX PRO 5000 Blackwell 48GB 1,344 GB/s. At 128k context both run the same 60 models we track fully in memory. On the models both run fully in memory, the NVIDIA RTX PRO 5000 Blackwell 48GB is about 1.4× as fast as the NVIDIA RTX 6000 Ada Generation (estimated, median across models).

Updated . Speeds are estimates; how they are worked out.

NVIDIA RTX 6000 Ada Generation

Not priced here

Add to build
NVIDIA RTX PRO 5000 Blackwell 48GB

Not priced here

Add to build

Specifications

SpecNVIDIA RTX 6000 Ada GenerationNVIDIA RTX PRO 5000 Blackwell 48GB
GPURTX 6000 Ada GenerationRTX PRO 5000 Blackwell 48GB
Memory48 GB GDDR648 GB GDDR7
Memory bus384-bit384-bit
Memory bandwidth960 GB/s1,344 GB/s
FP16 compute91.1 TFLOPS73.69 TFLOPS
SoftwareCUDACUDA
Board power300 W300 W
PCIe generationGen 4Gen 5
PCIe lanesx16x16

What each runs

Estimated decode speed with the most precise quant that fits each one est.

ModelNVIDIA RTX 6000 Ada GenerationNVIDIA RTX PRO 5000 Blackwell 48GB
Qwen 3.8 27B16tok/sQ8_022tok/sQ8_0
Qwen 3.5 122B-A10B53tok/sUD-Q2_K_XL74tok/sUD-Q2_K_XL
Qwen 3.6 27B16tok/sQ8_022tok/sQ8_0
Gemma 4 31B14tok/sQ8_019tok/sQ8_0
Mistral Small 4 119B66tok/sUD-Q2_K_XL93tok/sUD-Q2_K_XL
Qwen 3.6 35B-A3B61tok/sQ8_086tok/sQ8_0
Qwen 3.5 27B16tok/sQ8_022tok/sQ8_0
Gemma 4 26B-A4B51tok/sQ8_071tok/sQ8_0
Qwen 3.5 35B-A3B61tok/sQ8_086tok/sQ8_0
GLM 4.7 Flash29tok/sQ8_040tok/sQ8_0
Devstral Small 2 24B13tok/sQ8_018tok/sQ8_0
Nemotron 3.5 Lightning 30B-A3B86tok/sUD-Q6_K_XL120tok/sUD-Q6_K_XL
Kimi Linear 48B-A3B100tok/sQ6_K140tok/sQ6_K
gpt-oss 20B63tok/sMXFP489tok/sMXFP4
Qwen3 Coder 30B-A3B22tok/sQ8_031tok/sQ8_0
Gemma 4 12B39tok/sQ8_055tok/sQ8_0
Qwen 3.5 9B43tok/sQ8_060tok/sQ8_0
Gemma 3 27B15tok/sQ8_021tok/sQ8_0
Mistral Small 3.2 24B13tok/sQ8_018tok/sQ8_0
Ministral 3 14B17tok/sQ8_023tok/sQ8_0
Qwen3 14B16tok/sQ8_022tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B21tok/sQ8_030tok/sQ8_0
Qwen3 8B21tok/sQ8_030tok/sQ8_0
Phi-4 14B14tok/sQ8_020tok/sQ8_0
Qwen 3.5 4B67tok/sQ8_094tok/sQ8_0
Ministral 3 8B22tok/sQ8_031tok/sQ8_0
Granite 4.2 8B19tok/sQ8_027tok/sQ8_0
Gemma 4 E4B50tok/sQ8_070tok/sQ8_0
LFM2.5 8B-A1B48tok/sQ8_067tok/sQ8_0
Llama 3.1 8B25tok/sQ6_K35tok/sQ6_K
Qwen2.5 0.5B278tok/sQ8_0389tok/sQ8_0
Qwen3.5 0.8B244tok/sQ8_0341tok/sQ8_0
Qwen3.5 2B194tok/sQ5_K_M272tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B105tok/sQ8_0147tok/sQ8_0
Mellum2.1 12B A2.5B Thinking78tok/sQ8_0109tok/sQ8_0
Spark X2.5 4B75tok/smlx-4bit105tok/smlx-4bit
MiniCPM5 2B72tok/sQ8_0100tok/sQ8_0
FrogNano 4B 260966tok/sQ8_093tok/sQ8_0
LFM2.5 2.6B55tok/sQ8_076tok/sQ8_0
LFM2.5 VL 3B55tok/sQ8_076tok/sQ8_0
Kimi VL A3B Thinking 250650tok/sQ8_070tok/sQ8_0
Kimi VL A3B49tok/smlx-8bit68tok/smlx-8bit
GLM 4 9B 041444tok/smlx-5bit61tok/smlx-5bit
granite vision 4.1 4b42tok/sQ8_058tok/sQ8_0
granite 4.2 3b41tok/sQ8_057tok/sQ8_0
GLM 4.6V Flash39tok/sQ8_054tok/sQ8_0
DeepSeek R1 Distill Qwen 7B38tok/sQ8_053tok/sQ8_0
GLM 4.1V 9B Thinking35tok/smlx-8bit49tok/smlx-8bit
Ministral 3 3B Instruct 251234tok/sQ8_047tok/sQ8_0
LFM2.5 24B A2B28tok/sQ8_039tok/sQ8_0
DeepSeek R1 Distill Llama 8B23tok/sQ8_032tok/sQ8_0
Ministral 3 8B Reasoning 251222tok/sQ8_031tok/sQ8_0
Qwen3.8 27B OBLITERATED20tok/smlx-4bit28tok/smlx-4bit
Muse Glimmer 30B17tok/sQ8_024tok/sQ8_0
Ministral 3 14B Reasoning 251217tok/sQ8_023tok/sQ8_0
DeepSeek R1 Distill Qwen 14B14tok/sQ8_020tok/sQ8_0
Devstral Small 250713tok/sQ8_018tok/sQ8_0
Magistral Small 250613tok/sQ8_018tok/sQ8_0
Magistral Small 250713tok/sQ8_018tok/sQ8_0
Magistral Small 250913tok/sQ8_018tok/sQ8_0
Qwen3 Coder Nextoffloadoffload
Nemotron 3 Super 120B-A12Boffloadoffload
gpt-oss 120Boffloadoffload
GLM 4.5 Airoffloadoffload
Qwen3 Next 80B-A3Boffloadoffload
Qwen3 32Boffloadoffload
DeepSeek R1 Distill Qwen 32Boffloadoffload
granite 4.2 30boffloadoffload
DeepSeek V4 Flashwon't runwon't run
GLM 5.3 Flashwon't runwon't run
Qwen 3.8 Flash Nextwon't runwon't run
MiniMax M2.7won't runwon't run
Llama 3.3 70Bwon't runwon't run
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Distill Llama 70Bwon't runwon't run
DeepSeek R1 Zerowon't runwon't run
DeepSeek V3won't runwon't run
DeepSeek V3 0324won't runwon't run
DeepSeek V3.1won't runwon't run
DeepSeek V3.1 Terminuswon't runwon't run
DeepSeek V3.2won't runwon't run
DeepSeek V4 Flash 0731won't runwon't run
DeepSeek V4 Flash Vision Expwon't runwon't run
Devstral 2 123B Instruct 2512won't runwon't run
GLM 4.5won't runwon't run
GLM 4.5Vwon't runwon't run
GLM 4.6won't runwon't run
GLM 4.6Vwon't runwon't run
GLM 4.7won't runwon't run
GLM 5won't runwon't run
GLM 5.1won't runwon't run
GLM 5.2won't runwon't run
GLM 5.3won't runwon't run
Kimi Dev 72Bwon't runwon't run
Kimi K2won't runwon't run
Kimi K2 Instruct 0905won't runwon't run
Kimi K2 Thinkingwon't runwon't run
Kimi K2.5won't runwon't run
MiniMax M2won't runwon't run
MiniMax M2.1won't runwon't run
MiniMax M2.5won't runwon't run
MiniMax M3won't runwon't run
Mistral Medium 3.5 128Bwon't runwon't run

Questions

Which is better for local AI, the NVIDIA RTX 6000 Ada Generation or the NVIDIA RTX PRO 5000 Blackwell 48GB?
Choose the NVIDIA RTX PRO 5000 Blackwell 48GB for speed.
Which can run bigger models, the NVIDIA RTX 6000 Ada Generation or the NVIDIA RTX PRO 5000 Blackwell 48GB?
At 128k context both run the same 60 models we track fully in memory.
Which is faster for local AI, the NVIDIA RTX 6000 Ada Generation or the NVIDIA RTX PRO 5000 Blackwell 48GB?
On the models both run fully in memory, the NVIDIA RTX PRO 5000 Blackwell 48GB is about 1.4× as fast as the NVIDIA RTX 6000 Ada Generation (estimated, median across models).

More comparisons

More hardware for local AI