Skip to content

NVIDIA GeForce RTX 5070 Ti

NVIDIA · 16GB GDDR7 · Can run 61 models

Buy Amazon
Manufacturer NVIDIA
VRAM 16 GB
Memory Type GDDR7
Architecture Blackwell
CUDA Cores 8,960
Tensor Cores 280
Bandwidth 896 GB/s
TDP 300W
MSRP $749
Released Mar 10, 2025

AI Notes

The RTX 5070 Ti offers an impressive 896 GB/s bandwidth with 16GB GDDR7 VRAM. It runs 13B models at very fast speeds and can handle 30B models with quantization. The combination of high bandwidth and 16GB capacity makes it one of the best cards for local AI at its price point.

Compatible Models

Model Parameters Best Quant VRAM Used Fit Est. Speed
Qwen 3 0.6B 600M Q4_K_M 2.5 GB Runs ~358 tok/s
Qwen 3.5 0.8B 800M Q4_K_M 1.5 GB Runs ~597 tok/s
Gemma 3 1B 1B Q8_0 2 GB Runs ~448 tok/s
Llama 3.2 1B 1B Q8_0 3 GB Runs ~299 tok/s
DeepSeek R1 1.5B 1.5B Q8_0 3 GB Runs ~299 tok/s
Gemma 2 2B 2B Q8_0 4 GB Runs ~224 tok/s
Gemma 3n E2B 2B Q4_K_M 3.3 GB Runs ~272 tok/s
Gemma 4 E2B 2B Q4_K_M 4 GB Runs ~224 tok/s
Qwen 3.5 2B 2B Q4_K_M 3 GB Runs ~299 tok/s
Llama 3.2 3B 3B Q8_0 5 GB Runs ~179 tok/s
Phi-3 Mini 3.8B 3.8B Q8_0 5.8 GB Runs ~154 tok/s
Phi-4 Mini 3.8B 3.8B Q4_K_M 4.5 GB Runs ~199 tok/s
Gemma 3 4B 4B Q4_K_M 5 GB Runs ~179 tok/s
Gemma 3n E4B 4B Q4_K_M 4.5 GB Runs ~199 tok/s
Gemma 4 E4B 4B Q4_K_M 6 GB Runs ~149 tok/s
Qwen 3 4B 4B Q4_K_M 4.5 GB Runs ~199 tok/s
Qwen 3.5 4B 4B Q4_K_M 4.5 GB Runs ~199 tok/s
DeepSeek R1 7B 7B Q8_0 9 GB Runs ~100 tok/s
Falcon 3 7B 7B Q4_K_M 6.8 GB Runs ~132 tok/s
Mistral 7B 7B Q8_0 9 GB Runs ~100 tok/s
Qwen 2.5 7B 7B Q8_0 9 GB Runs ~100 tok/s
Qwen 2.5 Coder 7B 7B Q8_0 9 GB Runs ~100 tok/s
Qwen 2.5 VL 7B 7B Q4_K_M 7 GB Runs ~128 tok/s
Aya Expanse 8B 8B Q4_K_M 6.5 GB Runs ~138 tok/s
Cogito 8B 8B Q4_K_M 7.5 GB Runs ~119 tok/s
DeepSeek R1 8B 8B Q4_K_M 7.5 GB Runs ~119 tok/s
Llama 3.1 8B 8B Q8_0 10 GB Runs ~90 tok/s
Nemotron 3 Nano 8B 8B Q4_K_M 7.5 GB Runs ~119 tok/s
Qwen 3 8B 8B Q4_K_M 7.5 GB Runs ~119 tok/s
Gemma 2 9B 9B Q8_0 11 GB Runs ~81 tok/s
Qwen 3.5 9B 9B Q4_K_M 7.5 GB Runs ~119 tok/s
Falcon 3 10B 10B Q4_K_M 8.5 GB Runs ~105 tok/s
Llama 3.2 Vision 11B 11B Q4_K_M 8.5 GB Runs ~105 tok/s
Gemma 3 12B 12B Q4_K_M 10.5 GB Runs ~85 tok/s
Mistral Nemo 12B 12B Q4_K_M 9.5 GB Runs ~94 tok/s
DeepSeek R1 14B 14B Q4_K_M 9.9 GB Runs ~91 tok/s
Phi-4 14B 14B Q4_K_M 9.9 GB Runs ~91 tok/s
Phi-4 Reasoning 14B 14B Q4_K_M 11 GB Runs ~81 tok/s
Qwen 2.5 14B 14B Q4_K_M 9.9 GB Runs ~91 tok/s
Qwen 2.5 Coder 14B 14B Q4_K_M 12 GB Runs ~75 tok/s
Qwen 3 14B 14B Q4_K_M 12 GB Runs ~75 tok/s
Qwen 3.5 35B A3B 35B Q4_K_M 12 GB Runs ~75 tok/s
Codestral 22B 22B Q4_K_M 14.7 GB Runs (tight) ~61 tok/s
StarCoder2 15B 15B Q8_0 17 GB CPU Offload ~16 tok/s
Devstral 24B 24B Q4_K_M 17 GB CPU Offload ~16 tok/s
Magistral Small 24B 24B Q4_K_M 17 GB CPU Offload ~16 tok/s
Mistral Small 3.1 24B 24B Q4_K_M 18 GB CPU Offload ~15 tok/s
Gemma 4 26B 26B Q4_K_M 20 GB CPU Offload ~14 tok/s
Gemma 2 27B 27B Q4_K_M 17.7 GB CPU Offload ~15 tok/s
Gemma 3 27B 27B Q4_K_M 20 GB CPU Offload ~14 tok/s
Qwen 3.5 27B 27B Q4_K_M 19 GB CPU Offload ~14 tok/s
Qwen 3 30B-A3B (MoE) 30B Q4_K_M 22 GB CPU Offload ~12 tok/s
Gemma 4 31B 31B Q4_K_M 22 GB CPU Offload ~12 tok/s
Aya Expanse 32B 32B Q4_K_M 22 GB CPU Offload ~12 tok/s
Cogito 32B 32B Q4_K_M 21.5 GB CPU Offload ~13 tok/s
DeepSeek R1 32B 32B Q4_K_M 20.7 GB CPU Offload ~13 tok/s
Qwen 2.5 32B 32B Q4_K_M 20.7 GB CPU Offload ~13 tok/s
Qwen 2.5 Coder 32B 32B Q4_K_M 23 GB CPU Offload ~12 tok/s
Qwen 3 32B 32B Q4_K_M 23 GB CPU Offload ~12 tok/s
QwQ 32B 32B Q4_K_M 21.5 GB CPU Offload ~13 tok/s
Command R 35B 35B Q4_K_M 22.5 GB CPU Offload ~12 tok/s
23 model(s) are too large for this hardware.