LM Studio Slower Than Raw llama.cpp by ~40% on Qwen
MkGod · reddit · 2026-08-24
User benchmarked Qwen3.8-27B on dual RTX 5060 Ti, finding a significant performance gap between LM Studio and raw llama-server.exe.
Performance Delta:
- LM Studio: 30–40 tok/s.
- Raw llama-server.exe: 50–55 tok/s (40% faster).
- Settings were identical (TP, MTP speculative decoding), ruling out VRAM spillover.
Analysis:
- Suspects GUI wrapper overhead, context shift, or KV cache fragmentation in LM Studio.
- Despite a warning about CPU sampling fallback in SPLITMODETENSOR, raw llama.cpp remains significantly faster.
Reasoning Effort:
- lmstudio-community GGUF shows full Reasoning Effort dropdown.
- Unsloth GGUF shows only a basic On/Off toggle.
- Question: Is this controlled by GGUF metadata/Jinja templates or hardcoded LM Studio logic?
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