GPUs / NVIDIA
RTX 4070 Ti SUPER
- Memory
- GDDR6X
- Bus
- 256-bit
- Published peak
- 672 GB/s
- Capacity
- 16 GB
Measured
- Read
- 547.6GB/s
- Write
- 525.5GB/s
- Copy
- 540.3GB/s
- Random
- 83.8GB/s
- Latency
- 231ns
- Sustain
- 545.6GB/s
Medians of 2 verified runs from 1 operator. Read reaches 81 % of the published peak.
Local AI speed
| Model | Parameters | 4-bit, tokens/s | 8-bit, tokens/s |
|---|---|---|---|
| Llama 3.1 8B | 8.03B | ≤ 136 | ≤ 68 |
| Qwen 2.5 14B | 14.7B | ≤ 75 | ≤ 37 |
| gpt-oss-20b | 21B, 3.6B active | ≤ 304 | Does not fit |
| Mistral Small 24B | 24B | ≤ 46 | Does not fit |
| Qwen 2.5 32B | 32.5B | Does not fit | Does not fit |
| Llama 3.3 70B | 70.6B | Does not fit | Does not fit |
| gpt-oss-120b | 117B, 5.1B active | Does not fit | Does not fit |