NVIDIA GTX 1080 - 8GB usable
8GB Pascal with a little more bandwidth than the 1070 Ti and the same limits. Fine for 7B at Q4; anything larger disappoints.
Specifications
| Brand | NVIDIA |
|---|---|
| Model | GTX 1080 |
| Usable VRAM | 8GB |
| Architecture | Pascal |
| CUDA / Stream Processors | 2,560 |
| Memory Bandwidth | 320 GB/s |
| TDP | 180W |
| FP32 TFLOPS | 8.9 |
Current Offers
Used from $52New from $160
Prices last updated:
GPUDojo is reader-supported. When you buy through links on our site, we may earn an affiliate commission.
Price History
- eBay$52
- Amazon$157current
- Newegg$369mid-range
For AI / LLM Use
Limited VRAM restricts you to 7B quantized models. Slower generation, usable but not snappy. Older architecture may have limited software support (check CUDA compatibility).
What Models Can It Run?
- 7B Q6_K, 14B Q3_K (tight)
- 7B Q4_K_M only
Estimated Performance
Generation: ~24 tokens/sec
Prefill: ~159 tokens/sec
Recommended Quantisations
- Q4_K_M for 7B models
- Q3_K for larger experiments
Pros & Cons
Pros
- Consumer card: easy to install, display output
Cons
- Only 8GB usable VRAM: limited to small models
- Moderate memory bandwidth: not the fastest for inference
- Older Pascal architecture: verify current CUDA support
Community Verdict
No community reviews yet for the GTX 1080. Know a good review? Let us know.