12GB · AI Score 2.9/100 · anchored estimate vs 51 measured cards
2.9 AI Score Includes estimates
We have not run NVIDIA TITAN V on our bench. These figures are anchored estimates, interpolated per workload against the 51 GPUs we did measure. On Llama 3.1 8B (Q4_K_M) NVIDIA TITAN V should deliver about 104.1 tokens/sec. Llama 3.3 70B does not fit. It needs roughly 42GB and this card has 12GB. For image generation, SDXL should run near 2.22 it/s, while FLUX.1-dev won't fit at BF16 (needs ~26GB). 8 of the 12 workloads won't fit on 12GB at the tested precision, Qwen3 32B, Llama 3.3 70B, Z-Image Turbo, FLUX.1-dev and others. We publish those as hard gates rather than quietly dropping to a smaller quant.
| Workload | Result | Telemetry | Data |
|---|---|---|---|
| MiniCPM5 2B | 192.05 tok/s | 106 W62°CQ4_K_M | ✓ Measured |
| LFM2.5 2.6B | 210.37 tok/s | 99 W62°CQ4_K_M | ✓ Measured |
| Granite 4.1 3B | 146.45 tok/s | 118 W61°CQ4_K_M | ✓ Measured |
| Nemotron 3 Nano 4B | 149.84 tok/s | 115 W62°CQ4_K_M | ✓ Measured |
| Qwen3-4B | 141.06 tok/s | 108 W59°CQ4_K_M | ✓ Measured |
| DeepSeek Coder 7B Instruct v1.5 | 114.27 tok/s | 128 W62°CQ4_K_M | ✓ Measured |
| Llama 3 8B | 102.27 tok/s | 136 W63°CQ4_K_M | ✓ Measured |
| Llama-3.1-8B | 102.28 tok/s | 136 W62°CQ4_K_M | ✓ Measured |
| Qwen3 8B | 98.34 tok/s | 135 W63°CQ4_K_M | ✓ Measured |
| Nemotron Nano 9B v2 | 75.62 tok/s | 133 W62°CQ4_K_M | ✓ Measured |
| Ornith 1.5 9B | 85.44 tok/s | 133 W63°CQ4_K_M | ✓ Measured |
| Gemma 4 12B | 59.29 tok/s | 133 W63°CQ4_K_M | ✓ Measured |
| Qwen2.5-Coder-14B | 56.25 tok/s | 145 W63°CQ4_K_M | ✓ Measured |
| Qwen3 14B | 57.47 tok/s | 145 W64°CQ4_K_M | ✓ Measured |
| Gemma 4 26B A4B | ✕ Won't fit needs ~20 GB | VRAM-gated at this precision | Est. |
| Qwen3 30B A3B | ✕ Won't fit needs ~20 GB | VRAM-gated at this precision | Est. |
| Gemma 4 31B | ✕ Won't fit needs ~22 GB | VRAM-gated at this precision | Est. |
| Qwen3 32B | ✕ Won't fit | VRAM-gated at this precision | Est. |
| Llama 3.3 70B | ✕ Won't fit | VRAM-gated at this precision | Est. |
| Workload | Result | Telemetry | Data |
|---|---|---|---|
| Stable Diffusion XL | 4.44 images/min | estimated | Est. |
| FLUX.1 dev | ✕ Won't fit | VRAM-gated at this precision | Est. |
| Z-Image Turbo | ✕ Won't fit | VRAM-gated at this precision | Est. |
| Workload | Result | Telemetry | Data |
|---|---|---|---|
| FLUX.1 Kontext dev | ✕ Won't fit | VRAM-gated at this precision | Est. |
| Qwen-Image-Edit | ✕ Won't fit | VRAM-gated at this precision | Est. |
| Workload | Result | Telemetry | Data |
|---|---|---|---|
| LTX-Video (distilled) | ✕ Won't fit | VRAM-gated at this precision | Est. |
| Wan 2.2 5B (720p) | ✕ Won't fit | VRAM-gated at this precision | Est. |
| Architecture | Volta (GV100) |
| CUDA cores | 5,120 |
| VRAM | 12GB HBM2 |
| Memory bus | 3072-bit |
| Memory bandwidth | 652.8 GB/s |
| Boost clock | 1,455 MHz |
| TDP | 250 W |
| Process | 12nm |
| Interface | PCIe 3.0 x16 |
| Release date | 2017-12-07 |
| Launch MSRP | $2,999 |
NVIDIA TITAN V scores 2.9/100, #62 of 102. It ran 4 of 12; 8 exceeded its 12GB. Figures are anchored estimates, not measurements, we flag that on every row.
100% = this card, AI & Machine Learning headline metric (AI Score). #23 of 61 desktop cards in this vertical.
| GPU | Relative | % | AI Score |
|---|---|---|---|
| NVIDIA GeForce RTX 4070 Ti | 110% | 3.2 | |
| GeForce RTX 5070 | 110% | 3.2 | |
| NVIDIA GeForce RTX 4070 Super | 107% | 3.1 | |
| NVIDIA GeForce RTX 4070 | 107% | 3.1 | |
| NVIDIA TITAN V | 100% | 2.9 | |
| NVIDIA GeForce RTX 3060 | 90% | 2.6 | |
| AMD Radeon RX 7900 XT | 86% | 2.5 | |
| AMD Radeon RX 7800 XT | 83% | 2.4 | |
| AMD Radeon RX 9070 XT | 83% | 2.4 |
Same card, other workloads: NVIDIA TITAN V Gaming benchmarks
← All AI & Machine Learning GPU rankings
| Transistors | 21,100 million |
| Die size | 815 mm² |
| Process node | 12 nm |
| Fabricated by | TSMC |
| Transistor density | 25.9 million per mm² |
Denser than 74% of the 746 cards we have silicon data for. Density is the clearest measure of what a process node bought: a card that gained it without growing the die got its speed from the fab rather than the architecture.
Silicon figures from Wikipedia (CC BY-SA 4.0). Benchmarks on this page are our own. Compare every chip.
Whole-job timings, composed from our measured per-model results on this card.
| Workflow | Time | Energy | Basis |
|---|---|---|---|
| Full codebase review | 17.8 min | 42.81 Wh | measured |
Can't run: Product photo shoot (needs FLUX.1 Kontext dev), Photo restoration batch (needs FLUX.1 Kontext dev), Restore and enlarge photos (needs FLUX.1 Kontext dev), Character sheet, 12 poses (needs FLUX.1 dev), Short social clips (needs Qwen3 32B), 60-second AI short film (needs Qwen3 32B), 24-frame storyboard (needs Qwen3 32B), 6-panel comic page (needs Qwen3 32B), Long-form article batch (needs Llama 3.3 70B).
This card is $2,999 to buy. The cheapest listed rate on Vast.ai is $0.136/hour, but that is the floor: we budget $0.163/hour, a 20% premium, because idle time, storage and unavailable cheap instances all land on the same bill. At that rate buying wins after 18,376 GPU-hours. Below it you are paying for idle silicon.
| How you would use it | GPU-hours a year | Rental cost a year | Time to break even |
|---|---|---|---|
| 2 hours a day, hobby | 730 | $119 | 25.2 years |
| 8 hours a day, working on it | 2,920 | $477 | 6.3 years |
| 24/7, always-on agent | 8,760 | $1,430 | 2.1 years |
At hobby usage this card is very unlikely to pay for itself before it is superseded. Rent it. Rental figures include a 20% premium over the cheapest listed rate. Ignores electricity, resale and the fact that a rented card can be a newer one tomorrow.
Cheapest of the RunPod and Vast on-demand rates we see, sampled daily. Spot and interruptible pricing runs lower.