Hands-on: Qwen 3.8 27B achieves 100 tok/s on RTX 4090
doodlestein · x · 2026-08-25
A user shared their experience running Qwen 3.8 27B Q4 on an RTX 4090, achieving a generation speed of 100 tok/s using Multi-Token Prediction (MTP). When tasked with a complex Rust project including a GUI, the model successfully delivered a working result after compacting the context twice. The user rated the experience as comparable to top-tier commercial models, noting its high offline productivity potential.
Related event: Qwen 3.8 27B Hits 100 tok/s on a Single RTX 4090(3 posts)→
More from Models
- Muse Spark vs Qwen and GLM: US open models still trail on agentic tasks — zsakib_ · 2026-10-16
- Muse Spark 816B gets llama.cpp support just 30 minutes after release — ngxson · 2026-10-16
- Are the fastest GPU kernels now written by AI? Kernels are hyper-verifiable, argues one dev — fleetwood___ · 2026-10-16
- Scale AI launches muse spark open: 816B-param MoE with 1M context, billed as top US open-weight model — jiayi_pirate · 2026-10-16
- NVIDIA HF listing for Qwen3.8-Flash-Next-FP4-Local leaks unannounced Qwen3.8 model — TheZachMueller · 2026-10-16
- Muse Spark Open, billed as the most capable US open-weight model, lands on Nebius day one — Arindam_1729 · 2026-10-16