RX 6800 vs RX 7700 XT for local AI
The RX 6800 fits more: up to gpt-oss 20B, against Qwen3 14B on the RX 7700 XT. The RX 6800 generates about 19% faster on any model both can run, because it has 512 GB/s of memory bandwidth to the RX 7700 XT's 432 GB/s.
Memory and bandwidth
Memory decides which models fit. Bandwidth decides how fast they generate, because every token reads the model's active weights once.
| RX 6800 | RX 7700 XT | |
|---|---|---|
| Memory | 16 GB VRAM | 12 GB VRAM |
| Bandwidth | 512 GB/s | 432 GB/s |
| Biggest model that fits | gpt-oss 20B | Qwen3 14B |
| Best everyday pick | gpt-oss 20B, ~90 tok/s | Qwen3 14B, ~26 tok/s |
Every model, side by side
Estimated generation speed at the largest memory size each was sold in (16 GB and 12 GB).
| Model | RX 6800 | RX 7700 XT |
|---|---|---|
| Qwen3 4B About 4 GB | ~110 tok/s Runs great | ~95 tok/s Runs great |
| Qwen3 8B About 7 GB | ~54 tok/s Runs great | ~46 tok/s Runs great |
| Gemma 3 12B About 10 GB | ~35 tok/s Runs great | ~29 tok/s Usable |
| Qwen3 14B About 11 GB | ~30 tok/s Runs great | ~26 tok/s Usable |
| gpt-oss 20B About 16 GB | ~90 tok/s Runs great | Needs 16 GB |
| Gemma 3 27B About 19 GB | Needs 19 GB | Needs 19 GB |
| Qwen3 30B-A3B About 21 GB | Needs 21 GB | Needs 21 GB |
| Qwen3 32B About 22 GB | Needs 22 GB | Needs 22 GB |
| Llama 3.3 70B About 45 GB | Needs 45 GB | Needs 45 GB |
| gpt-oss 120B About 67 GB | Needs 67 GB | Needs 67 GB |
| Qwen3 235B-A22B About 144 GB | Needs 144 GB | Needs 144 GB |
| DeepSeek R1 671B About 406 GB | Needs 406 GB | Needs 406 GB |
The RX 6800: it fits bigger models and runs faster. Price matters as much as either number: check used prices on the builds page, where every parts list has a measured or estimated speed.
Both columns use the same formula: memory bandwidth divided by the weights read per token, calibrated on our own rack, with Ollama's default 4-bit sizes. Real results are usually within about a third. Measured either one? Send us the number.