Can I Run Qwen3-Coder on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)?
Written by Jakub Rusinowski · Last updated September 29, 2026
Yes
Yes — Qwen3-Coder 30B-A3B (MoE) at Q6_K needs about 26.6 GB of the 32 GB usable on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM) (~5.4 GB spare), at ~219.3 tok/s (estimated), with room for about 32,768 tokens of context.
Confidence: high · Recommended quantization: Q6_K · Estimated speed: ~219.3 tok/s
See what else this hardware can run →
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
RTX 5090 Desktop (32 GB VRAM, 64 GB RAM) — what it gives a model
| Usable memory for models | 32 GB |
| Memory bandwidth | 1792 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
Qwen3-Coder on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM): memory by quantization
| Quant | Memory needed | Fits 32 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 62.6 GB | ✗ No | — | — | 61 GB |
| Q8_0 | 34 GB | ✗ No | — | — | 32.4 GB |
| Q6_K | 26.6 GB | ✓ Yes | 32K | ~219.3 tok/s | 25 GB |
| Q5_K_M | 23.2 GB | ✓ Yes | 64K | ~233.3 tok/s | 21.6 GB |
| Q4_K_M | 20 GB | ✓ Yes | 64K | ~248.3 tok/s | 18.4 GB |
| Q3_K_M | 14.6 GB | ✓ Yes | 128K | ~278.4 tok/s | 13 GB |
| Q2_K | 11.6 GB | ✓ Yes | 128K | ~298.3 tok/s | 10 GB |
Which Qwen3-Coder sizes fit
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| Qwen3-Coder 480B-A35B (MoE) | 291.6 GB | ✗ Too large | — |
| Qwen3-Coder-Next (80B-A3B MoE) | 51.6 GB | ✗ Too large | — |
| Qwen3-Coder 30B-A3B (MoE) | 20 GB | ✓ Fits | ~248.3 tok/s |
| Qwen3-Coder 8B | 6.8 GB | ✓ Fits | ~159.6 tok/s |
What to watch out for
- 2 larger variants of Qwen3-Coder do not fit and would need CPU offload or different hardware.
RTX 5090 desktop limitations
- 575 W board power — budget for a 1000 W+ PSU and the heat it puts into the room.
- Models larger than 32 GB must offload to system RAM, which costs roughly an order of magnitude in speed.
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.
- 32 GB of VRAM on the NVIDIA GeForce RTX 5090 at 1792 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 Qwen3-Coder on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)?
Yes — Qwen3-Coder 30B-A3B (MoE) at Q6_K needs about 26.6 GB of the 32 GB usable on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM) (~5.4 GB spare), at ~219.3 tok/s (estimated), with room for about 32,768 tokens of context.
Which quantization of Qwen3-Coder should I use on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)?
Q6_K — it needs about 26.6 GB of the 32 GB available, downloads as roughly 25 GB, and runs at an estimated 219.3 tokens/sec with up to 32K of context.
What limits Qwen3-Coder on RTX 5090 Desktop (32 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 5090 Desktop (32 GB VRAM, 64 GB RAM)
- SmolLM2 on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
- SmolLM3 on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
- StarCoder 2 on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
- Aya 3B (Tiny Aya) on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
- VibeThinker on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
Qwen3-Coder on GPUs
- Qwen3-Coder on NVIDIA GeForce RTX 5070
- Qwen3-Coder on NVIDIA GeForce RTX 5060 Ti 8GB
- Qwen3-Coder on NVIDIA GeForce RTX 5060
- Qwen3-Coder on NVIDIA GeForce RTX 4070 Ti
What This Model Is Good At
Model & Tools
← Can I Run It? | Qwen3-Coder model page | Check your hardware