§01·spec · /gpus
RTX 3090
nvidia30 series24GB VRAM
35 open-weights AI models run on the RTX 3090 — see which fit, how fast they go, and the VRAM each needs.
§02·models that run on this GPU
35 totalLLM · 15
| Model | Best speed | Min VRAM | Works | Evidence | |
|---|---|---|---|---|---|
| Qwen3 30B-A3B | 153.6tokens/s | 24GB | ✓ | 2benchesrecipe | check ↗ |
| gpt-oss 20B | 147.5tokens/s | 24GB | ✓ | 2benchesrecipe | check ↗ |
| Qwen3-8B | 115.3tokens/s | 24GB | ✓ | 2benchesrecipe | check ↗ |
| Qwen3 14B | 70tokens/s | 24GB | ✓ | 2benchesrecipe | check ↗ |
| Qwen3 32B | 35.1tokens/s | 24GB | ✓ | 2benchesrecipe | check ↗ |
| DeepSeek R1 Distill 14B | 12GB | ✓ | recipe | check ↗ | |
| Devstral Small 2 (24B) | 24GB | ✓ | recipe | check ↗ | |
| Gemma 4 12B | 24GB | ✓ | recipe | check ↗ | |
| Laguna XS 2.1 | 24GB | ✓ | recipe | check ↗ | |
| Llama 3.1 8B | 10GB | ✓ | recipe | check ↗ | |
| Mistral Nemo 12B | 24GB | ✓ | recipe | check ↗ | |
| Mistral Small 3.2 24B | 24GB | ✓ | recipe | check ↗ | |
| North Mini Code 1.0 | 24GB | ✓ | recipe | check ↗ | |
| Ornith 1.0 35B | 24GB | ✓ | recipe | check ↗ | |
| Phi-4 | 24GB | ✓ | recipe | check ↗ |
Multimodal · 5
| Model | Best speed | Min VRAM | Works | Evidence | |
|---|---|---|---|---|---|
| Gemma 4 26B MoE | 119.4tokens/s | 24GB | ✓ | 2benchesrecipe | check ↗ |
| Qwen3.5 35B | 111.2tokens/s | 24GB | ✓ | 2benchesrecipe | check ↗ |
| Gemma4 31B | 34.7tokens/s | 24GB | ✓ | 2benchesrecipe | check ↗ |
| Qwen3.5 27B | 33.5tokens/s | 24GB | ✓ | 2benchesrecipe | check ↗ |
| Gemma 4 E4B-IT | 20GB | ✓ | recipe | check ↗ |
Image · 8
| Model | Best speed | Min VRAM | Works | Evidence | |
|---|---|---|---|---|---|
| Krea 2 | 14.75s/img | 18.75GB | ✓ | 1benchrecipe | check ↗ |
| Z-Image Turbo | 9s | 13GB | ✓ | 1benchrecipe | check ↗ |
| Chroma V48 | 24GB | ✓ | recipe | check ↗ | |
| Flux.2-Klein-4B | 13GB | ✓ | recipe | check ↗ | |
| HiDream-O1-Image | 20GB | ✓ | recipe | check ↗ | |
| Juggernaut Z | 16GB | ✓ | recipe | check ↗ | |
| LongCat Image | 18GB | ✓ | recipe | check ↗ | |
| Qwen-Image | 13GB | ✓ | recipe | check ↗ |
Video · 6
3D · 1
| Model | Best speed | Min VRAM | Works | Evidence | |
|---|---|---|---|---|---|
| Waypoint 1.5 | 12GB | ✓ | recipe | check ↗ |
§03·tested recipes
showing 6 of 35- llmintermediate24GB+recipe
Gemma 4 12B on RTX 3090: Local Private Assistant via llama.cpp / Ollama (24GB)
- llmintermediate24GB+recipe
Phi-4 (14B) on RTX 3090: Local Private Assistant via llama.cpp / Ollama (24GB)
- llmintermediate24GB+recipe
Mistral Nemo 12B on RTX 3090: Local Private Assistant via llama.cpp / Ollama (24GB)
- llmintermediate24GB+recipe
Mistral Small 3.2 24B on RTX 3090: Local Private Assistant via llama.cpp / Ollama (24GB)
- llmintermediate24GB+recipe
Devstral Small 2 (24B) on RTX 3090: Local Agentic Coding via llama.cpp + OpenHands (24GB Value Tier)
- llmadvanced24GB+recipe
Laguna XS 2.1 on RTX 3090: Local Agentic Coding via Ollama / llama.cpp + OpenHands (24GB Value-Entry Tier)