NVIDIA RTX 3070 Ti
GPUAmpere; 8GB VRAM, 608 GB/s bandwidth, 290W TDP, CUDA cores 6,144. Release date: 2021-06-10. Source: LCZ hardware catalog, with missing fields supplemented from NVIDIA official materials.
2
Run records
1
Models covered
1
Frameworks covered
L0 Self-reported × 2
Evidence mix
Typical measured performance
Published benchmark averages per model × framework × quantization on this hardware.
- llama.cpp · Q5_K_M Decode 158.1 · Prefill 1185.8 ·
- llama.cpp · Q4_K_M Decode 197.6 · Prefill 1482 ·
Key specs
- Vendor
- NVIDIA
- Architecture
- Ampere
- VRAM
- 8 GB
- Memory bandwidth
- 608 GB/s
- TDP
- 290 W
- Release date
- 2021-06-10
Run records
2 configurations
| Model | Framework | Quantization | Decode | Gen time | Samples |
|---|---|---|---|---|---|
| Qwen3.5-4B | llama.cpp | Q5_K_M | 158.1 tok/s | — | 1 |
| Qwen3.5-4B | llama.cpp | Q4_K_M | 197.6 tok/s | — | 1 |