Supported models for local AI
Browse Ora-supported local models and enter your hardware to see what your machine can run privately.
Your hardware
Enter available VRAM, system RAM, and free storage to filter the catalog.
| Model | Min VRAM | Min RAM | Storage | Products | Status |
|---|---|---|---|---|---|
Qwen3 Coder A3B Qwen · 30B | 2 GB | 8 GB | 15 GB | CLI | Coming soon |
Qwen3 Coder Next Qwen · 80B | 4 GB | 8 GB | 40 GB | CLI | Coming soon |
GPT OSS GPT · 120B | 4 GB | 8 GB | 70 GB | CLI | Coming soon |
DeepSeek V4 Flash DeepSeek · 284B | 8 GB | 16 GB | 140 GB | CLI | Coming soon |
Requirements are minimum estimates for Ora-optimized artifacts. Actual performance depends on your GPU, drivers, and workload.
