What can I run with 16 GB of VRAM?
With 16 GB of VRAM (or unified memory) and an 8K-token conversation, 8 of the 16 language models we track fit comfortably, and the largest is gpt-oss-20b.
Language models (LLMs)
| Model | Best precision | Memory | Verdict |
|---|---|---|---|
| gpt-oss-20bOpenAI | 16-bit (FP16/BF16) | 14.0 GB | Fits |
| Qwen2.5 14B InstructAlibaba Qwen | 4-bit (Q4_K_M) | 10.6 GB | Fits |
| Qwen3 14BAlibaba Qwen | 4-bit (Q4_K_M) | 10.3 GB | Fits |
| DeepSeek-R1-0528-Qwen3-8BDeepSeek | 8-bit (Q8_0) | 9.9 GB | Fits |
| Llama 3.1 8B InstructMeta | 8-bit (Q8_0) | 9.6 GB | Fits |
| Qwen2.5 7B InstructAlibaba Qwen | 8-bit (Q8_0) | 8.6 GB | Fits |
| Mistral 7B Instruct v0.3Mistral AI | 8-bit (Q8_0) | 8.8 GB | Fits |
| Llama 3.2 3B InstructMeta | 16-bit (FP16/BF16) | 7.4 GB | Fits |
| Qwen3.8 Flash NextAlibaba Qwen | needs 119.6 GB at 4-bit | 119.6 GB* | Too big |
| gpt-oss-120bOpenAI | needs 63.1 GB at 4-bit | 63.1 GB | Too big |
| Llama 3.1 70B InstructMeta | needs 45.2 GB at 4-bit | 45.2 GB | Too big |
| Mixtral 8x7B Instruct v0.1Mistral AI | needs 27.5 GB at 4-bit | 27.5 GB | Too big |
| Qwen2.5 32B InstructAlibaba Qwen | needs 22.0 GB at 4-bit | 22.0 GB | Too big |
| Gemma 4 31BGoogle | needs 19.6 GB at 4-bit | 19.6 GB* | Too big |
| Qwen3.8 27BAlibaba Qwen | needs 18.0 GB at 4-bit | 18.0 GB | Too big |
| Gemma 4 26B A4BGoogle | needs 17.0 GB at 4-bit | 17.0 GB* | Too big |
Memory needed = weights at that precision (real Hugging Face download sizes) + FP16 KV cache for 8,192 tokens. "Fits" means under 90% of 16 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 | Works, tight |
| 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.