Running Qwen3.8-Flash-Next FP8 at 524K context on dual RTX 6000
SpendLucky1273 · reddit · 2026-08-29
The author successfully ran Qwen3.8-Flash-Next FP8 at 524K context on 2x RTX PRO 6000 using vLLM.
Config: TP2 + EP2, MTP3, PLE CPU offload, YaRN 2x, and --max-model-len 524288.
Issue & Fix:
- When extending beyond 262K, the target model updated correctly, but the MTP draft remained at 262K, causing validation failures.
- Fix: Patched the Qwen3.8 MTP config to inherit the target maxmodellen before vLLM builds the cache.
Status: Boots cleanly with 654,980 tokens KV cache and 1.25x concurrency.
More from Infra
- SK Hynix ties new US plant to looser chip equipment exports to China, sources say — pstAsiatech · 2026-10-15
- Perplexity's new embedding models: index with a 9B model, search with 0.6B — antoine_chaffin · 2026-10-15
- Unsloth ships GGUF quants for Nemotron 3.5: Bolt-74B runs in 24GB RAM — danielhanchen · 2026-10-15
- vLLM-Omni ships Day-0 serving for FLUX 3 Image Klein with BF16, FP8, NVFP4 support — vllm_project · 2026-10-15
- NVIDIA's Nemotron 3.5 Super VL lands training and inference support on Applied Compute — ypatil125 · 2026-10-15
- CoreWeave bond yield tops 16% as loans outlast customer contracts by ~2 years — tengyanAI · 2026-10-15