RTX 5090 meets the minimum VRAM requirement for Q4 inference of 01-ai/Yi-1.5-34B-Chat. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
RTX 5090 can run 01-ai/Yi-1.5-34B-Chat with Q4 quantization. At approximately 99 tokens/second, you can expect Good speed - acceptable for interactive use.
You have 14GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 18GB | 32GB | 98.80 tok/s | ✅ Fits comfortably |
| Q8 | 35GB | 32GB | 74.97 tok/s | ❌ Not recommended |
| FP16 | 70GB | 32GB | 37.04 tok/s | ❌ Not recommended |