§01·compatibility · /check
Qwen3-8B on RTX 5090
Yes — Qwen3-8B runs on the RTX 5090 (32 GB). Fastest community-measured result: 11933.4 prefill tokens/s.
✓ runsllmactive50 series32GB VRAM
§02·benchmarks
| Task | Quant | Speed | VRAM | Works | Confidence | Source | Verified |
|---|---|---|---|---|---|---|---|
| llm | Q4_K | 11933.4prefill tokens/s | ✓ | hardware-corner.net· web | 2026-05-15 | ||
| llm | Q4_K | 200.4tokens/s | ✓ | hardware-corner.net· web | 2026-05-15 | ||
| llm | Q4_K_XL | 145.34tokens/s | ✓ | hardware-corner.net· web | 2026-05-15 |
§03·common questions
Can you run Qwen3-8B on RTX 5090?
Yes — Qwen3-8B runs on the RTX 5090 (32 GB). Fastest community-measured result: 11933.4 prefill tokens/s.
Which quantizations have been tested for Qwen3-8B on RTX 5090?
Q4_K, Q4_K_XL — measured in community benchmarks.
How fast is Qwen3-8B on RTX 5090?
Up to 11933.4 prefill tokens/s (llm), the fastest community-measured result.
Are there step-by-step instructions for Qwen3-8B on RTX 5090?
Yes — a step-by-step recipe documents Qwen3-8B on the RTX 5090; see the recipes listed below.
§04·related recipes
- llmbeginner6GB+
Qwen3-8B on RTX 5090: Q4_K_M GGUF with 26 GB of Headroom for Colocation, BF16, or Full 131K Context
- llmbeginner5GB+
Qwen3-8B on Apple M2 Pro: local 8B chat with MLX 4-bit in 16 GB unified memory
- llmbeginner9GB+
Qwen3-8B on RX 7800 XT: 16 GB ROCm via Ollama or llama.cpp-HIP GGUF
- llmbeginner6GB+
Qwen3-8B on RX 7900 XTX: ROCm via Ollama or llama.cpp-HIP
- llmbeginner12GB+
Qwen3-8B on RTX 3080 Ti 12GB: Q4_K_M GGUF via Ollama or llama.cpp