What can I run with 128 GB of VRAM?
With 128 GB of VRAM (or unified memory) and an 8K-token conversation, 15 of the 16 language models we track fit comfortably, and the largest is gpt-oss-120b.
Language models (LLMs)
| Model | Best precision | Memory | Verdict |
|---|---|---|---|
| gpt-oss-120bOpenAI | 16-bit (FP16/BF16) | 65.7 GB | Fits |
| Llama 3.1 70B InstructMeta | 8-bit (Q8_0) | 77.7 GB | Fits |
| Mixtral 8x7B Instruct v0.1Mistral AI | 16-bit (FP16/BF16) | 94.5 GB | Fits |
| Qwen2.5 32B InstructAlibaba Qwen | 16-bit (FP16/BF16) | 67.7 GB | Fits |
| Gemma 4 31BGoogle | 16-bit (FP16/BF16) | 61.4 GB* | Fits |
| Qwen3.8 27BAlibaba Qwen | 16-bit (FP16/BF16) | 55.2 GB | Fits |
| Gemma 4 26B A4BGoogle | 16-bit (FP16/BF16) | 50.5 GB* | Fits |
| gpt-oss-20bOpenAI | 16-bit (FP16/BF16) | 14.0 GB | Fits |
| Qwen2.5 14B InstructAlibaba Qwen | 16-bit (FP16/BF16) | 31.2 GB | Fits |
| Qwen3 14BAlibaba Qwen | 16-bit (FP16/BF16) | 30.9 GB | Fits |
| DeepSeek-R1-0528-Qwen3-8BDeepSeek | 16-bit (FP16/BF16) | 17.6 GB | Fits |
| Llama 3.1 8B InstructMeta | 16-bit (FP16/BF16) | 17.1 GB | Fits |
| Qwen2.5 7B InstructAlibaba Qwen | 16-bit (FP16/BF16) | 15.7 GB | Fits |
| Mistral 7B Instruct v0.3Mistral AI | 16-bit (FP16/BF16) | 15.6 GB | Fits |
| Llama 3.2 3B InstructMeta | 16-bit (FP16/BF16) | 7.4 GB | Fits |
| Qwen3.8 Flash NextAlibaba Qwen | 4-bit (Q4_K_M) | 119.6 GB* | Tight |
Memory needed = weights at that precision (real Hugging Face download sizes) + FP16 KV cache for 8,192 tokens. "Fits" means under 90% of 128 GB; "Tight" means 4-bit loads but leaves under 10% free. A * means the KV cache isn't sourced yet, so only the weights are counted. Data checked 2026-10-23.
Image generation, video editing and gaming
| Workload | Min–comfortable | Verdict |
|---|---|---|
| Stable Diffusion 1.5 (512×512)Image generation | 4–8 GB | Comfortable |
| SDXL 1.0 (1024×1024)Image generation | 8–12 GB | Comfortable |
| FLUX.1 [dev] (1024×1024)Image generation | 8–16 GB | Comfortable |
| Video editing, 1080p (HD)Video editing | 4–8 GB | Comfortable |
| Video editing, 4KVideo editing | 8–12 GB | Comfortable |
| Video editing, 8KVideo editing | 16–24 GB | Comfortable |
| Gaming, 1080pGaming | 8–12 GB | Comfortable |
| Gaming, 1440pGaming | 12–16 GB | Comfortable |
| Gaming, 4KGaming | 16–16 GB | Comfortable |
Ranges come from published guidance (Adobe, Puget Systems, AUTOMATIC1111 and others); the sources are linked in the calculator when you pick a workload.
Other memory sizes
8 GB12 GB16 GB24 GB32 GB48 GB64 GB96 GB128 GB
Want a different context length, more than one GPU, or your exact card? Use the What can my GPU run? tool. To check one model in detail, use the VRAM calculator.