Can I Run OLMo 2 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?
作者： Jakub Rusinowski · 最后更新： 2024年11月26日
Yes, but it is tight
Yes, but it is tight — OLMo 2 13B Instruct at Q8_0 needs about 22.1 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM), leaving only ~1.9 GB before the runtime starts swapping. Expect ~38.6 tok/s (estimated), with room for about 4,096 tokens of context.
Confidence: medium · Recommended quantization: Q8_0 · Estimated speed: ~38.6 tok/s
Get a personalized upgrade path →
或在 Vast.ai 比较，低至 $0.35/小时 (typical low · varies)
作为亚马逊联盟成员，我们从符合条件的购买中获得收入。云 GPU 链接为推荐链接——我们可能获得佣金，您无需额外付费。
RTX 4090 Desktop (24 GB VRAM, 64 GB RAM) — what it gives a model
| Usable memory for models | 24 GB |
| Memory bandwidth | 1008 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
OLMo 2 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM): memory by quantization
| Quant | Memory needed | Fits 24 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 34.9 GB | ✗ No | — | — | 27.4 GB |
| Q8_0 | 22.1 GB | ✓ Yes | 4K | ~38.6 tok/s | 14.6 GB |
| Q6_K | 18.7 GB | ✓ Yes | 4K | ~46.4 tok/s | 11.2 GB |
| Q5_K_M | 17.2 GB | ✓ Yes | 4K | ~51.2 tok/s | 9.7 GB |
| Q4_K_M | 15.8 GB | ✓ Yes | 4K | ~56.7 tok/s | 8.3 GB |
| Q3_K_M | 13.4 GB | ✓ Yes | 4K | ~69.4 tok/s | 5.8 GB |
| Q2_K | 12 GB | ✓ Yes | 4K | ~79.1 tok/s | 4.5 GB |
Which OLMo 2 sizes fit
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| OLMo 2 13B Instruct | 15.8 GB | ✓ Fits | ~56.7 tok/s |
| OLMo 2 7B Instruct | 9.5 GB | ✓ Fits | ~91.6 tok/s |
What to watch out for
- Context is capped at about 4,096 tokens before memory runs out, which is short for document or agent work.
RTX 4090 desktop limitations
- 24 GB is the sweet spot for 27–32B models at Q4; 70B needs offload or a second card.
Recommended setup
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
How these numbers are calculated
- Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
- 24 GB of VRAM on the NVIDIA GeForce RTX 4090 at 1008 GB/s.
- 64 GB of system RAM available for CPU offload when a model exceeds VRAM.
- Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
- KV cache is computed from this model's published attention configuration.
FAQ
Can I run OLMo 2 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?
Yes, but it is tight — OLMo 2 13B Instruct at Q8_0 needs about 22.1 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM), leaving only ~1.9 GB before the runtime starts swapping. Expect ~38.6 tok/s (estimated), with room for about 4,096 tokens of context.
Which quantization of OLMo 2 should I use on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?
Q8_0 — it needs about 22.1 GB of the 24 GB available, downloads as roughly 14.6 GB, and runs at an estimated 38.6 tokens/sec with up to 4K of context.
What limits OLMo 2 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Which runtime should I use?
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Other Computers
Other Models on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- Phi 3.5 Family on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- Phi-4 Family on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- Phi-4 Mini on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- Poolside Laguna XS 2.1 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- Qwen 2.5 Family on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
OLMo 2 on GPUs
- OLMo 2 on NVIDIA GeForce RTX 5080
- OLMo 2 on NVIDIA GeForce RTX 5070 Ti
- OLMo 2 on NVIDIA GeForce RTX 5060 Ti 16GB
- OLMo 2 on NVIDIA GeForce RTX 5060 Ti 8GB