StarCoder 2 15B on NVIDIA GeForce RTX 4070 Ti 12GB
NVIDIA GeForce RTX 4070 Ti 12GB can run StarCoder 2 15B at 28 tok/s at Q3_K_M, though performance is acceptable. Consider a higher-end GPU for better results.
Model Size
15.5B
Device VRAM
12 GB
Bandwidth
504 GB/s
Quantizations Tested
1
Performance by Quantization
Each row shows StarCoder 2 15B performance at a different quality level on NVIDIA GeForce RTX 4070 Ti 12GB.
| Quantization | Speed | TTFT | Fits in VRAM | Rating | Confidence |
|---|---|---|---|---|---|
| Q3_K_M | 28 tok/s | 220ms | ✓ Yes | Acceptable | estimated |
Notes
Q3_K_M
Q3_K_M fits. Q4 would need ~10.5GB — tight. Compressed but usable for coding.
About StarCoder 2 15B
StarCoder 2 15B (15.5B) is a coding, ai coding model. Trained on The Stack v2 with 619 programming languages. Strong fill-in-the-middle (FIM) support for code completion. BigCode's best model — competitive but surpassed by Qwen 2.5 Coder on most benchmarks.
View all StarCoder 2 15B hardware options →About NVIDIA GeForce RTX 4070 Ti 12GB
NVIDIA GeForce RTX 4070 Ti 12GB has 12 GB at 504 GB/s. Street price: $749.
See all models NVIDIA GeForce RTX 4070 Ti 12GB can run →Source: 15.73B model. Q3_K_M fits with headroom (2026-03-15)
Data last updated: 2026-03-01