This GPU offers reliable throughput for local AI workloads. Pair it with the right model quantization to hit your desired tokens/sec, and monitor prices below to catch the best deal.
Quick Answer: NVIDIA A100 40GB PCIe has 40GB VRAM, enough for models up to roughly 100B parameters at 4-bit quantization. It draws 250W under load.
With 40GB VRAM, NVIDIA A100 40GB PCIe can run models up to approximately 100B parameters using 4-bit quantization. That covers most popular models, including 70B-class ones at 4-bit.
Consider RTX 4090 or RTX 6000 Ada — 24GB Ada offers better efficiency than Ampere.
Buy directly on Amazon with fast shipping and reliable customer service.
Essential accessories to pair with NVIDIA A100 40GB PCIe
Accessories Total
Typical prices — check Amazon for current pricing
💡 Not ready to buy? Try cloud GPUs first
Test NVIDIA A100 40GB PCIe performance in the cloud before investing in hardware. Pay by the hour with no commitment.
Showing 12 of 80 rows. Speeds are calculated estimates, not measurements — search for your model to jump straight to it.
| Model | Size | Quantization | Tokens/sec | VRAM used |
|---|---|---|---|---|
| Deepseek AI Deepseek Coder 1.3B Instruct | 1.3B | Q4 | ~340 tok/sEstimated | 1GB |
| Deepseek AI Deepseek R1 Distill Qwen 1.5B | 1.5B | Q4 | ~340 tok/sEstimated | 1GB |
| Deepseek AI Deepseek Ocr 2 | Unknown | Q4 | ~285 tok/sEstimated | 2GB |
| Deepseek AI Deepseek Ocr | Unknown | Q4 | ~285 tok/sEstimated | 2GB |
| Lmstudio Community Deepseek R1 0528 Qwen3 8B Mlx 8bit | 8B | Q4 | ~285 tok/sEstimated | 4GB |
| Lmstudio Community Deepseek R1 0528 Qwen3 8B Mlx 4bit | 8B | Q4 | ~285 tok/sEstimated | 4GB |
| Deepseek AI Deepseek R1 Distill Qwen 7B | 7B | Q4 | ~285 tok/sEstimated | 4GB |
| Nineninesix Kani Tts 2 En | Unknown | Q4 | ~275 tok/sEstimated | 1GB |
| Qwen Qwen3 Tts 12hz 1 7B Customvoice | 7B | Q4 | ~275 tok/sEstimated | 1GB |
| Zai Org Glm Ocr | Unknown | Q4 | ~275 tok/sEstimated | 1GB |
| Qwen Qwen3 Asr 1 7B | 7B | Q4 | ~275 tok/sEstimated | 2GB |
| Nari Labs Dia2 2B | 2B | Q4 | ~275 tok/sEstimated | 1GB |
Showing 12 of 240 rows.
| Model | Size | Quantization | Verdict | Estimated speed | VRAM needed |
|---|---|---|---|---|---|
| 01 AI Yi 1 5 34B Chat | 34B | Q4 | Fits comfortably | ~80 tok/sEstimated | 18GB (have 40GB) |
| 01 AI Yi 1 5 34B Chat | 34B | Q8 | Fits comfortably | ~56 tok/sEstimated | 35GB (have 40GB) |
| 01 AI Yi 1 5 34B Chat | 34B | FP16 | Not supported | ~30 tok/sEstimated | 69GB (have 40GB) |
| AI Forever Rugpt 3.5 13B | 13B | Q4 | Fits comfortably | ~170 tok/sEstimated | 7GB (have 40GB) |
| AI Forever Rugpt 3.5 13B | 13B | Q8 | Fits comfortably | ~120 tok/sEstimated | 13GB (have 40GB) |
| AI Forever Rugpt 3.5 13B | 13B | FP16 | Fits comfortably | ~65 tok/sEstimated | 26GB (have 40GB) |
| AI Mo Kimina Prover 72B | 72B | Q4 | Fits comfortably | ~45 tok/sEstimated | 37GB (have 40GB) |
| AI Mo Kimina Prover 72B | 72B | Q8 | Not supported | ~32 tok/sEstimated | 73GB (have 40GB) |
| AI Mo Kimina Prover 72B | 72B | FP16 | Not supported | ~17 tok/sEstimated | 146GB (have 40GB) |
| Alibaba Nlp Gte Qwen2 1.5B Instruct | 1.5B | Q4 | Fits comfortably | ~275 tok/sEstimated | 1GB (have 40GB) |
| Alibaba Nlp Gte Qwen2 1.5B Instruct | 1.5B | Q8 | Fits comfortably | ~190 tok/sEstimated | 2GB (have 40GB) |
| Alibaba Nlp Gte Qwen2 1.5B Instruct | 1.5B | FP16 | Fits comfortably | ~105 tok/sEstimated | 4GB (have 40GB) |
Note: Performance estimates are calculated. Real results may vary. Methodology · Submit real data
Explore how RTX 4090 stacks up for local inference workloads.
Explore how RTX 4080 stacks up for local inference workloads.
Explore how RTX 4070 Ti stacks up for local inference workloads.
Explore how RTX 3090 stacks up for local inference workloads.
Explore how RX 7900 XTX stacks up for local inference workloads.