NVIDIA GeForce RTX 4070 Ti
NVIDIA · 12GB GDDR6X · Can run 16 models
| Manufacturer | NVIDIA |
| VRAM | 12 GB |
| Memory Type | GDDR6X |
| Architecture | Ada Lovelace |
| CUDA Cores | 7,680 |
| Tensor Cores | 240 |
| TDP | 285W |
| MSRP | $799 |
| Released | Jan 5, 2023 |
AI Notes
The RTX 4070 Ti provides solid AI inference capability with 12GB of GDDR6X VRAM. It can run 7B-parameter models at full precision and 13B models with quantization. The 12GB VRAM limit means larger models require aggressive quantization or offloading to system RAM.
Compatible Models
| Model | Parameters | Best Quant | VRAM Used | Fit |
|---|---|---|---|---|
| Llama 3.2 1B | 1B | Q8_0 | 3 GB | Runs |
| Gemma 2 2B | 2B | Q8_0 | 4 GB | Runs |
| Llama 3.2 3B | 3B | Q8_0 | 5 GB | Runs |
| Phi-3 Mini 3.8B | 3.8B | Q8_0 | 5.8 GB | Runs |
| DeepSeek R1 7B | 7B | Q8_0 | 9 GB | Runs |
| Mistral 7B | 7B | Q8_0 | 9 GB | Runs |
| Qwen 2.5 7B | 7B | Q8_0 | 9 GB | Runs |
| Qwen 2.5 Coder 7B | 7B | Q8_0 | 9 GB | Runs |
| Llama 3.1 8B | 8B | Q8_0 | 10 GB | Runs |
| DeepSeek R1 14B | 14B | Q4_K_M | 9.9 GB | Runs |
| Phi-4 14B | 14B | Q4_K_M | 9.9 GB | Runs |
| Qwen 2.5 14B | 14B | Q4_K_M | 9.9 GB | Runs |
| Gemma 2 9B | 9B | Q8_0 | 11 GB | Runs (tight) |
| StarCoder2 15B | 15B | Q8_0 | 17 GB | CPU Offload |
| Codestral 22B | 22B | Q4_K_M | 14.7 GB | CPU Offload |
| Gemma 2 27B | 27B | Q4_K_M | 17.7 GB | CPU Offload |
9
model(s) are too large for this hardware.