Can RTX 4090 run Qwen3.8 27B?
Last updated:
Every measured result for Qwen3.8 27B on RTX 4090
| Build | Quant / setup | Engine | tok/s | |
|---|---|---|---|---|
| RTX 4090: Qwen3.8 27B Q4_K_M at 43.1 tok/s (llm-speed, llama.cpp b11803, 6 runs) | Q4_K_M | llama.cpp build 11803 (a37cc4610) | 43.1 | source |
| EPYC 7642 + RTX 4090 + 252 GB RAM: Qwen3.8 27B Q4_K_M at 43.1 tok/s (llm-speed community run) | Q4_K_M | llama.cpp (df18bee) | 43.06 | source |
| RTX 4090: Qwen3.8 27B Q4_K_M at 43.1 tok/s (llm-speed, llama.cpp b11803, 6 runs) | Q4_K_M | llama.cpp build 11803 (a37cc4610) | 43.0 | source |
| EPYC 7642 + RTX 4090 + 252 GB RAM: Qwen3.8 27B Q4_K_M at 43.1 tok/s (llm-speed community run) | Q4_K_M | llama.cpp (df18bee) | 42.63 | source |
Memory: RTX 4090 has 24 GB. Qwen3.8 27B weights are about 16.5 GB at Q4_K_M (27B params × 4.89 bits ÷ 8, before context/KV cache). Arithmetic, not a measurement; long contexts need more. Check other quants in the VRAM calculator.
Speeds are decode (generation) tokens/sec as reported by each source; quant, engine and context differ, so small gaps are noise.
All measured speeds for Qwen3.8 27B: best home rig for Qwen3.8 27B.
Qwen3.8 27B on other hardware: 2x Apple M5 Ultra · 2x Intel Arc Pro B70 · 2x RTX 3090 · 2x RTX 5070 Ti · 2x Tesla V100 32GB · 4x Apple M5 Ultra · Apple M3 Ultra · Apple M5 Ultra · DGX Spark (GB10) · Intel Arc Pro B70 · RTX 3060 12GB · RTX 3090 · RTX 4060 Ti 16GB · RTX 5060 Ti 16GB · RTX 5070 Ti · RTX 5090 · RTX Pro 6000 Max-Q (Blackwell, 96 GB, 300 W) · RTX Spark N1X · Radeon AI Pro R9700 · Strix Halo (Radeon 8060S)
Other models measured on RTX 4090: Flint 40B-A2.5B · GLM 4.7 Flash · Gemma 4 12B · Gemma 4 26B-A4B · Gemma 4 31B · LFM2.5 24B-A2B · Muse Glimmer 30B · Nemotron 3.5 Bolt 74B-A9B · Qwen3.5 9B · Qwen3.6 27B · Qwen3.6 35B-A3B · gpt-oss 20B
Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms). It never affects our data.