RTX 2060 6GBBudget pickC191.5 tok/s decode
~$349 MSRP
Can it run?
Tight fit
Using Q4_K_M in Ollama
Fit status
Tight fit
Decode
64.0 tok/s
TTFT
3025 ms
Safe context
19K
Memory
3.3 GB / 4.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Tight fit | 64.0 tok/s | 4400 ms | 39K |
| Chat | C | Tight fit | 64.0 tok/s | 1650 ms | 10K |
| Coding | C | Tight fit | 64.0 tok/s | 3025 ms | 19K |
| RAG | C | Tight fit | 64.0 tok/s | 5500 ms | 39K |
| Reasoning | C | Tight fit | 64.0 tok/s | 3575 ms | 19K |
How Yi Coder 1.5B (1.5B params) fits at each quantization level on GTX 1650 4GB (4.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 0.6 GB | Low | D32 |
Q3_K_S | 3 | 0.7 GB | Low | D33 |
NVFP4 | 4 | 0.8 GB | Medium | D33 |
Q4_K_M | 4 | 0.9 GB | Medium | D34 |
Q5_K_M | 5 | 1.1 GB | High | D35 |
Q6_K | 6 | 1.2 GB | High | D35 |
Q8_0Best for your GPU | 8 | 1.6 GB | Very High | D38 |
F16 | 16 | 3.1 GB | Maximum | C44 |
Upgrade options
RTX 2060 6GBBudget pick~$349 MSRP
Intel Arc A380 6GBBiggest leap
GTX 1060 6GBNVIDIA upgrade