community pool / model answer
Qwen-AgentWorld-35B-A3B-GGUF
chat · unsloth/Qwen-AgentWorld-35B-A3B-GGUF
69.8 tok/s · model median · 19 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| RTX 3060 12GB ×2 | 4-bit | 78.3 tok/s | measured | - | 21.6 GB | 32,768 | llama.cpp |
| Ryzen AI Max 395 128GB | 4-bit | 34.9 tok/s | reported | 37,267 | 21.8 GB | 65,536 | llama.cpp |
| Ryzen AI Max 395 64GB | 4-bit | 60.5 tok/s | measured | 1,576 | 21.5 GB | 8,192 | llama.cpp |
Context tested
The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.