Qwen model 14x slower on 4070 due to layers offloading to RAM
luckokkkk · reddit · 2026-08-24
A user discovered that Qwen3.8-27B ran at a sluggish 5.7 tok/s on an RTX 4070 Super, compared to 81.5 tok/s on an RTX 5090.
Investigation revealed:
- While the model has 66 layers, only 38 ran on the 4070 Super's GPU. The remaining 28 layers offloaded to system RAM due to VRAM constraints, creating a massive bottleneck.
Solution:
- The author released Picchio, an open-source tool that visualizes actual layer placement, GPU activity, and separate prefill/decode speeds in a single run, helping developers debug inference performance issues effectively.
More from Infra
- engy.ai processes 50B+ tokens in a day as open-model inference demand surges — const_reborn · 2026-10-16
- New inference engine 'e' hits 10,500 tok/s on a single DGX Station — brandon_galang · 2026-10-16
- SGLang ships Day-0 support for FLUX 3 Image Klein: 10 reference images, BF16/FP8/NVFP4 variants — ying11231 · 2026-10-16
- NVIDIA's Nemotron 3.5 Super VL hits CoreWeave: 12B-active multimodal MoE at $0.25/M input tokens — wandb · 2026-10-16
- Chip equipment beats bull case: 2026 WFE seen at $153B, $278B by 2030 — BenBajarin · 2026-10-16
- Steam now launches in 5.23s, beating Epic's 5.47s — and vLLM gets roasted for slow startup — AlpinDale · 2026-10-16