Written by Jakub Rusinowski · Last updated July 15, 2026
Yes, but it is tight
Yes, but it is tight — 1-bit Bonsai 27B at Q5_K_M needs about 21.7 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM), leaving only ~2.3 GB before the runtime starts swapping. Expect ~34.8 tok/s (estimated), with room for about 16,384 tokens of context.
Confidence: low · Recommended quantization: Q5_K_M · Estimated speed: ~34.8 tok/s
| Usable memory for models | 24 GB |
| Memory bandwidth | 1008 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
| Quant | Memory needed | Fits 24 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 56.5 GB | ✗ No | — | — | 54 GB |
| Q8_0 | 31.2 GB | ✗ No | — | — | 28.7 GB |
| Q6_K | 24.7 GB | ✗ No | — | — | 22.1 GB |
| Q5_K_M | 21.7 GB | ✓ Yes | 16K | ~34.8 tok/s | 19.1 GB |
| Q4_K_M | 18.8 GB | ✓ Yes | 16K | ~40.1 tok/s | 16.3 GB |
| Q3_K_M | 14.1 GB | ✓ Yes | 32K | ~53.7 tok/s | 11.5 GB |
| Q2_K | 11.4 GB | ✓ Yes | 64K | ~66 tok/s | 8.9 GB |
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| 1-bit Bonsai 27B | 18.8 GB | ✓ Fits | ~40.1 tok/s |
| Ternary Bonsai 27B | 18.8 GB | ✓ Fits | ~40.1 tok/s |
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Yes, but it is tight — 1-bit Bonsai 27B at Q5_K_M needs about 21.7 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM), leaving only ~2.3 GB before the runtime starts swapping. Expect ~34.8 tok/s (estimated), with room for about 16,384 tokens of context.
Q5_K_M — it needs about 21.7 GB of the 24 GB available, downloads as roughly 19.1 GB, and runs at an estimated 34.8 tokens/sec with up to 16K of context.
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
← Can I Run It? | Bonsai 27B model page | Check your hardware