readygpu

NVIDIA GeForce RTX 3090 vs NVIDIA GeForce RTX 5070

For running models locally, priced in Canada. At 32k context the first runs 48 models fully in memory, the second 23.

In short

The NVIDIA GeForce RTX 3090 has more memory for models: 24 GB against 12 GB. At 32k context the NVIDIA GeForce RTX 3090 runs 48 of the models we track fully in memory and the NVIDIA GeForce RTX 5070 23; only the NVIDIA GeForce RTX 3090 holds Qwen 3.8 27B. On the models both run fully in memory, the NVIDIA GeForce RTX 3090 is about 1.4× as fast as the NVIDIA GeForce RTX 5070 (estimated, median across models). In Canada the NVIDIA GeForce RTX 3090 costs $2,800 at Amazon and the NVIDIA GeForce RTX 5070 $1,181 at Amazon: the NVIDIA GeForce RTX 5070 is $1,619 cheaper.

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

NVIDIA GeForce RTX 3090

$2,800

Amazon

Add to build
NVIDIA GeForce RTX 5070

$1,181

Amazon

Add to build

Specifications

SpecNVIDIA GeForce RTX 3090NVIDIA GeForce RTX 5070
GPUGeForce RTX 3090GeForce RTX 5070
Memory24 GB GDDR6X12 GB GDDR7
Memory bus384-bit192-bit
Memory bandwidth936 GB/s672 GB/s
FP16 compute71 TFLOPS62 TFLOPS
SoftwareCUDACUDA
Board power350 W250 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 3090NVIDIA GeForce RTX 5070
Qwen 3.8 27B25tok/sQ5_K_Moffload
Qwen 3.6 27B27tok/sQ5_K_Moffload
Gemma 4 31B27tok/sQ4_K_Moffload
Qwen 3.6 35B-A3B134tok/sQ4_K_Moffload
Qwen 3.5 27B27tok/sQ5_K_Moffload
Gemma 4 26B-A4B78tok/sUD-Q6_Koffload
Qwen 3.5 35B-A3B135tok/sQ4_K_Moffload
GLM 4.7 Flash65tok/sQ5_K_Moffload
Devstral Small 2 24B26tok/sQ5_K_Moffload
gpt-oss 20B108tok/sMXFP4offload
Qwen3 Coder 30B-A3B69tok/sQ4_K_Moffload
Gemma 4 12B43tok/sQ8_039tok/sQ6_K
Qwen 3.5 9B55tok/sQ8_039tok/sQ8_0
Mistral Small 3.2 24B26tok/sQ5_K_Moffload
Gemma 3 27B26tok/sQ5_K_Moffload
Ministral 3 14B29tok/sQ8_0offload
Qwen3 14B28tok/sQ8_0offload
DeepSeek R1 0528 Qwen3 8B43tok/sQ8_039tok/sQ5_K_M
Qwen3 8B43tok/sQ8_039tok/sQ5_K_M
Phi-4 14B26tok/sQ8_0offload
Qwen 3.5 4B104tok/sQ8_074tok/sQ8_0
Ministral 3 8B43tok/sQ8_039tok/sQ5_K_M
Granite 4.2 8B39tok/sQ8_038tok/sQ4_K_M
Gemma 4 E4B63tok/sQ8_045tok/sQ8_0
LFM2.5 8B-A1B130tok/sQ8_094tok/sQ8_0
Llama 3.1 8B53tok/sQ6_K38tok/sQ6_K
Qwen2.5 0.5B622tok/sQ8_0446tok/sQ8_0
Qwen3.5 0.8B471tok/sQ8_0338tok/sQ8_0
Qwen3.5 2B313tok/sQ5_K_M224tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B205tok/sQ8_0147tok/sQ8_0
MiniCPM5 2B142tok/sQ8_0102tok/sQ8_0
LFM2.5 2.6B119tok/sQ8_085tok/sQ8_0
LFM2.5 VL 3B119tok/sQ8_085tok/sQ8_0
Mellum2.1 12B A2.5B Thinking110tok/sQ8_0offload
granite vision 4.1 4b92tok/sQ8_066tok/sQ8_0
granite 4.2 3b88tok/sQ8_063tok/sQ8_0
Kimi VL A3B81tok/smlx-8bitoffload
LFM2.5 24B A2B78tok/sQ6_Koffload
DeepSeek R1 Distill Qwen 7B58tok/sQ8_042tok/sQ8_0
GLM 4.6V Flash51tok/sQ8_056tok/sQ4_K_M
DeepSeek R1 Distill Llama 8B45tok/sQ8_038tok/sQ6_K
GLM 4.1V 9B Thinking44tok/smlx-8bitoffload
Ministral 3 8B Reasoning 251243tok/sQ8_039tok/sQ5_K_M
DeepSeek R1 Distill Qwen 14B26tok/sQ8_0offload
Devstral Small 250726tok/sQ5_K_Moffload
Magistral Small 250626tok/sQ5_K_Moffload
Magistral Small 250726tok/sQ5_K_Moffload
Magistral Small 250926tok/sQ5_K_Moffload
Qwen 3.5 122B-A10Boffloadwon't run
Qwen3 Coder Nextoffloadwon't run
Mistral Small 4 119Boffloadwon't run
GLM 4.5 Airoffloadwon't run
Qwen3 Next 80B-A3Boffloadwon't run
Nemotron 3.5 Lightning 30B-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 V3 0324won't runwon't run
DeepSeek V3.1won'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 4.5Vwon't runwon't run
GLM 4.6Vwon't runwon't run
GLM 4.7won'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 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 3090 or the NVIDIA GeForce RTX 5070?
Choose the NVIDIA GeForce RTX 3090 to run bigger models. Choose the NVIDIA GeForce RTX 3090 for speed. Choose the NVIDIA GeForce RTX 5070 to spend less.
Which can run bigger models, the NVIDIA GeForce RTX 3090 or the NVIDIA GeForce RTX 5070?
At 32k context the NVIDIA GeForce RTX 3090 runs 48 of the models we track fully in memory and the NVIDIA GeForce RTX 5070 23; only the NVIDIA GeForce RTX 3090 holds Qwen 3.8 27B.
Which is faster for local AI, the NVIDIA GeForce RTX 3090 or the NVIDIA GeForce RTX 5070?
On the models both run fully in memory, the NVIDIA GeForce RTX 3090 is about 1.4× as fast as the NVIDIA GeForce RTX 5070 (estimated, median across models).
What is the price difference between the NVIDIA GeForce RTX 3090 and the NVIDIA GeForce RTX 5070?
In Canada the NVIDIA GeForce RTX 3090 costs $2,800 at Amazon and the NVIDIA GeForce RTX 5070 $1,181 at Amazon: the NVIDIA GeForce RTX 5070 is $1,619 cheaper.

More comparisons