Stable Diffusion 1.5 VRAM requirements
The 2022 model that started it all. Under 1B parameters, native 512 px, and still a huge library of fine-tunes. Runs on almost anything.
TypeImage
Parameters0.98B
Fits entirely from4 GB
8-bit or better from4 GB
01The files, and how much VRAM each needs
| File | Size | Needed | Min. VRAM | Quality | Source |
|---|---|---|---|---|---|
| 16-bit | 2.1 GB | 3.3 GB | 4 GB | the original weights | Comfy-Org/stable-diffusion-v1-5-archive → |
“Needed” = file + 0.8 GB working memory + 0.8 GB system reserve. The 2.13 GB checkpoint also holds the text encoder and VAE; the UNet alone is about 1.7 GB at 16-bit.
02By amount of VRAM
03Best GPU for SD 1.5
The cheapest cards (by launch price) that run it well, and every card sorted by memory: best GPU for SD 1.5 → Planning bigger images or longer clips? Open the calculator →
04By graphics card
| GPU | VRAM | Verdict | Best file | Needed |
|---|---|---|---|---|
| Desktop graphics cards | ||||
| RTX 2060 6 GB | 6 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3050 6 GB | 6 GB | Runs well | 16-bit | 3.3 GB |
| RTX 2070 Super 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 2080 Super 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3050 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3060 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3060 Ti 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3070 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3070 Ti 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4060 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4060 Ti 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5050 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5060 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5060 Ti 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RX 7600 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RX 9050 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RX 9060 XT 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| Arc B570 10 GB | 10 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3080 10 GB | 10 GB | Runs well | 16-bit | 3.3 GB |
| RTX 2080 Ti 11 GB | 11 GB | Runs well | 16-bit | 3.3 GB |
| Arc B580 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 2060 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3060 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3080 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3080 Ti 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4070 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4070 Super 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4070 Ti 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5070 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RX 7700 XT 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RX 9070 GRE 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| Arc A770 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4060 Ti 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4070 Ti Super 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4080 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4080 Super 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5060 Ti 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5070 Ti 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5080 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RX 7600 XT 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RX 7800 XT 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RX 7900 GRE 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RX 9060 XT 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RX 9070 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RX 9070 XT 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RX 7900 XT 20 GB | 20 GB | Runs well | 16-bit | 3.3 GB |
| Arc Pro B60 24 GB | 24 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3090 24 GB | 24 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3090 Ti 24 GB | 24 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4090 24 GB | 24 GB | Runs well | 16-bit | 3.3 GB |
| RX 7900 XTX 24 GB | 24 GB | Runs well | 16-bit | 3.3 GB |
| Arc Pro B70 32 GB | 32 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5090 32 GB | 32 GB | Runs well | 16-bit | 3.3 GB |
| Laptop GPUs | ||||
| RTX 3050 Laptop 4 GB | 4 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3050 Ti Laptop 4 GB | 4 GB | Runs well | 16-bit | 3.3 GB |
| RTX 2060 Laptop 6 GB | 6 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3050 Laptop 6 GB | 6 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3060 Laptop 6 GB | 6 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4050 Laptop 6 GB | 6 GB | Runs well | 16-bit | 3.3 GB |
| RTX 2070 Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 2070 Super Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 2080 Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 2080 Super Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3070 Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3070 Ti Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3080 Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4060 Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4070 Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5050 Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5060 Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5070 Laptop 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RX 7600M 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RX 7600M XT 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RX 7600S 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RX 7700S 8 GB | 8 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4080 Laptop 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5070 Laptop 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5070 Ti Laptop 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RX 7800M 12 GB | 12 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3080 Laptop 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 3080 Ti Laptop 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 4090 Laptop 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5080 Laptop 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RX 7900M 16 GB | 16 GB | Runs well | 16-bit | 3.3 GB |
| RTX 5090 Laptop 24 GB | 24 GB | Runs well | 16-bit | 3.3 GB |
| Unified memory | ||||
| Radeon 8060S (Strix Halo) 96 GB | 96 GB | Runs well | 16-bit | 3.3 GB |
| Radeon 8065S (Gorgon Halo) 160 GB | 160 GB | Runs well | 16-bit | 3.3 GB |
| Apple Silicon Macs (by memory) | ||||
| Mac 16 GB | 12.7 GB | Runs well | 16-bit | 3.3 GB |
| Mac 18 GB | 14.4 GB | Runs well | 16-bit | 3.3 GB |
| Mac 24 GB | 19.6 GB | Runs well | 16-bit | 3.3 GB |
| Mac 32 GB | 26.8 GB | Runs well | 16-bit | 3.3 GB |
| Mac 36 GB | 30.2 GB | Runs well | 16-bit | 3.3 GB |
| Mac 48 GB | 40.2 GB | Runs well | 16-bit | 3.3 GB |
| Mac 64 GB | 55.7 GB | Runs well | 16-bit | 3.3 GB |
| Mac 96 GB | 85 GB | Runs well | 16-bit | 3.3 GB |
| Mac 128 GB | 115.4 GB | Runs well | 16-bit | 3.3 GB |
| Mac 192 GB | 175.4 GB | Runs well | 16-bit | 3.3 GB |
| Mac 256 GB | 236.9 GB | Runs well | 16-bit | 3.3 GB |
| Mac 512 GB | 498.1 GB | Runs well | 16-bit | 3.3 GB |
05Text encoder, VAE and other files
The text encoder is inside the checkpoint, so there is nothing extra to download.
06Where the files go in ComfyUI
| File | Folder | Loader node |
|---|---|---|
| The checkpoint (.safetensors) | ComfyUI/models/checkpoints | Load Checkpoint |
07Measured and reported results
No measured results yet. Send yours.
08Training a LoRA for SD 1.5
| Trainer | VRAM | Type | Settings and quote | Source |
|---|---|---|---|---|
| OneTrainer | 2.8 GB | example run | Community low-VRAM config (efhosci) on OneTrainer wiki: fp8 weights, batch 1, 512px, rank 16, SDP attention, gradient checkpointing, latent caching, AdamW8bit “With all above optimizations, the program maxed out around 2.8 GB while running.” | github.com → |
| OneTrainer | 6 GB | stated minimum | LoRA or embedding at 512px; general FAQ answer, no other settings given “SD1.5 LoRA or embedding with 512 resolution can be trained with just 6Gb.” | github.com → |
reported From each trainer's own docs, 2026-09-25. Why the figures differ →
09Every file tracked for SD 1.5
| File | Type | Size | Repo |
|---|---|---|---|
| v1-5-pruned-emaonly.safetensors | F32 | 4.27 GB | Comfy-Org/stable-diffusion-v1-5-archive → all-in-one checkpoint: includes text encoder(s) and VAE, not diffusion-model-only (CLIP-L + VAE); fp32 weights |
| v1-5-pruned-emaonly-fp16.safetensors | F16 | 2.13 GB | Comfy-Org/stable-diffusion-v1-5-archive → all-in-one checkpoint: includes text encoder(s) and VAE, not diffusion-model-only (CLIP-L + VAE) |
UNet ~0.86B + CLIP-L; native 512px. Only all-in-one checkpoints (Load Checkpoint node); no widely used ComfyUI GGUF of SD1.5 checked (not needed: whole checkpoint is ~2 GB at fp16). Huge fine-tune ecosystem (Civitai).