readygpu

NVIDIA GeForce RTX 4090 vs NVIDIA RTX PRO 4000 Blackwell

For running models locally, priced in Germany. At 32k context the first runs 57 models fully in memory, the second 57.

In short

Both have 24 GB of memory for models; the NVIDIA GeForce RTX 4090 has 1,008 GB/s of bandwidth and the NVIDIA RTX PRO 4000 Blackwell 672 GB/s. At 32k context both run the same 57 models we track fully in memory. On the models both run fully in memory, the NVIDIA GeForce RTX 4090 is about 1.5× as fast as the NVIDIA RTX PRO 4000 Blackwell (estimated, median across models). In Germany the NVIDIA GeForce RTX 4090 costs 3.915 € at eBay and the NVIDIA RTX PRO 4000 Blackwell 3.014 € at eBay: the NVIDIA RTX PRO 4000 Blackwell is 901 € cheaper.

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

NVIDIA GeForce RTX 4090

3.915 €

eBay

Add to build
NVIDIA RTX PRO 4000 Blackwell

3.014 €

eBay

Add to build

Specifications

SpecNVIDIA GeForce RTX 4090NVIDIA RTX PRO 4000 Blackwell
GPUGeForce RTX 4090RTX PRO 4000 Blackwell
Memory24 GB GDDR6X24 GB GDDR7
Memory bus384-bit192-bit
Memory bandwidth1,008 GB/s672 GB/s
FP16 compute165 TFLOPS46.9 TFLOPS
SoftwareCUDACUDA
Board power450 W145 W
PCIe generationGen 4Gen 5
PCIe lanesx16x16

What each runs

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

ModelNVIDIA GeForce RTX 4090NVIDIA RTX PRO 4000 Blackwell
Qwen 3.8 27B27tok/sQ5_K_M18tok/sQ5_K_M
Qwen 3.6 27B29tok/sQ5_K_M19tok/sQ5_K_M
Gemma 4 31B29tok/sQ4_K_M19tok/sQ4_K_M
Qwen 3.6 35B-A3B145tok/sQ4_K_M96tok/sQ4_K_M
Qwen 3.5 27B29tok/sQ5_K_M19tok/sQ5_K_M
Gemma 4 26B-A4B84tok/sUD-Q6_K56tok/sUD-Q6_K
Qwen 3.5 35B-A3B146tok/sQ4_K_M97tok/sQ4_K_M
GLM 4.7 Flash70tok/sQ5_K_M47tok/sQ5_K_M
Devstral Small 2 24B28tok/sQ5_K_M19tok/sQ5_K_M
Nemotron 3.5 Lightning 30B-A3B153tok/sMXFP4102tok/sMXFP4
gpt-oss 20B116tok/sMXFP477tok/sMXFP4
Qwen3 Coder 30B-A3B74tok/sQ4_K_M50tok/sQ4_K_M
Gemma 4 12B46tok/sQ8_031tok/sQ8_0
Qwen 3.5 9B59tok/sQ8_039tok/sQ8_0
Mistral Small 3.2 24B28tok/sQ5_K_M19tok/sQ5_K_M
Gemma 3 27B28tok/sQ5_K_M19tok/sQ5_K_M
Ministral 3 14B32tok/sQ8_021tok/sQ8_0
Qwen3 14B30tok/sQ8_020tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B46tok/sQ8_031tok/sQ8_0
Qwen3 8B46tok/sQ8_031tok/sQ8_0
Phi-4 14B28tok/sQ8_019tok/sQ8_0
Qwen 3.5 4B112tok/sQ8_074tok/sQ8_0
Ministral 3 8B46tok/sQ8_031tok/sQ8_0
Granite 4.2 8B43tok/sQ8_028tok/sQ8_0
Gemma 4 E4B68tok/sQ8_045tok/sQ8_0
LFM2.5 8B-A1B140tok/sQ8_094tok/sQ8_0
Llama 3.1 8B57tok/sQ6_K38tok/sQ6_K
Qwen2.5 0.5B669tok/sQ8_0446tok/sQ8_0
Qwen3.5 0.8B507tok/sQ8_0338tok/sQ8_0
Qwen3.5 2B337tok/sQ5_K_M224tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B221tok/sQ8_0147tok/sQ8_0
MiniCPM5 2B153tok/sQ8_0102tok/sQ8_0
Spark X2.5 4B145tok/smlx-4bit97tok/smlx-4bit
LFM2.5 2.6B128tok/sQ8_085tok/sQ8_0
LFM2.5 VL 3B128tok/sQ8_085tok/sQ8_0
Mellum2.1 12B A2.5B Thinking118tok/sQ8_079tok/sQ8_0
FrogNano 4B 2609109tok/sQ8_073tok/sQ8_0
granite vision 4.1 4b99tok/sQ8_066tok/sQ8_0
granite 4.2 3b95tok/sQ8_063tok/sQ8_0
Kimi VL A3B Thinking 250691tok/sQ8_061tok/sQ8_0
Kimi VL A3B88tok/smlx-8bit58tok/smlx-8bit
Ministral 3 3B Instruct 251288tok/sQ8_058tok/sQ8_0
LFM2.5 24B A2B84tok/sQ6_K56tok/sQ6_K
GLM 4 9B 041465tok/smlx-5bit43tok/smlx-5bit
DeepSeek R1 Distill Qwen 7B63tok/sQ8_042tok/sQ8_0
GLM 4.6V Flash55tok/sQ8_037tok/sQ8_0
DeepSeek R1 Distill Llama 8B49tok/sQ8_033tok/sQ8_0
GLM 4.1V 9B Thinking48tok/smlx-8bit32tok/smlx-8bit
Ministral 3 8B Reasoning 251246tok/sQ8_031tok/sQ8_0
Ministral 3 14B Reasoning 251232tok/sQ8_021tok/sQ8_0
DeepSeek R1 Distill Qwen 14B28tok/sQ8_019tok/sQ8_0
Devstral Small 250728tok/sQ5_K_M19tok/sQ5_K_M
Magistral Small 250628tok/sQ5_K_M19tok/sQ5_K_M
Magistral Small 250728tok/sQ5_K_M19tok/sQ5_K_M
Magistral Small 250928tok/sQ5_K_M19tok/sQ5_K_M
Qwen3.8 27B OBLITERATED27tok/smlx-4bit18tok/smlx-4bit
Muse Glimmer 30B26tok/sQ6_K17tok/sQ6_K
Qwen 3.5 122B-A10Boffloadoffload
Qwen3 Coder Nextoffloadoffload
Mistral Small 4 119Boffloadoffload
GLM 4.5 Airoffloadoffload
Qwen3 Next 80B-A3Boffloadoffload
Kimi Linear 48B-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
Nemotron 3 Super 120B-A12Bwon't runwon't run
gpt-oss 120Bwon'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 GeForce RTX 4090 or the NVIDIA RTX PRO 4000 Blackwell?
Choose the NVIDIA GeForce RTX 4090 for speed. Choose the NVIDIA RTX PRO 4000 Blackwell to spend less.
Which can run bigger models, the NVIDIA GeForce RTX 4090 or the NVIDIA RTX PRO 4000 Blackwell?
At 32k context both run the same 57 models we track fully in memory.
Which is faster for local AI, the NVIDIA GeForce RTX 4090 or the NVIDIA RTX PRO 4000 Blackwell?
On the models both run fully in memory, the NVIDIA GeForce RTX 4090 is about 1.5× as fast as the NVIDIA RTX PRO 4000 Blackwell (estimated, median across models).
What is the price difference between the NVIDIA GeForce RTX 4090 and the NVIDIA RTX PRO 4000 Blackwell?
In Germany the NVIDIA GeForce RTX 4090 costs 3.915 € at eBay and the NVIDIA RTX PRO 4000 Blackwell 3.014 € at eBay: the NVIDIA RTX PRO 4000 Blackwell is 901 € cheaper.

More comparisons

More hardware for local AI