sybil-solutions/qwen38-b70

Fastest validated sglang recipe for Qwen3.8-27B (AWQ-INT4 + DSpark) on RTX 3090s — 1×/2×/4× GPU presets, image+video, full benchmarks

Reproducible Qwen3.8-27B GPTQ INT4 + MTP4 inference on Intel Arc Pro B70 using vLLM XPU — 84.65 tok/s median.

Intel Arc LLM inference server (A770/B60): Qwen hybrids up to Qwen3.8-Flash-Next (512-expert MoE through an expert cache in VRAM and host RAM) on llama.cpp with arcint's own OpenCL kernels, MTP speculation, context checkpoints; image generation, vision, speech in and out

Deployment kit: Qwen3.8-27B EXL3 3.5bpw target + DFlash2 EXL3 5.0bpw speculative draft — launcher, config, model cards

A llama.cpp fork with TurboQuant/TCQ KV cache, fused Qwen MTP speculative decoding, and multi-GPU long-context serving fixes.

Qwen3.8-27B-GSQ3 on one RTX 4080: 100k context with rk4v4-e8, up to 2720 tok/s prefill and 262 tok/s generation

Hyper optimised Qwen3.8-27B inference on one RTX 3090: ReplaySSM, MTP3, reasoning effort, C1-C8 batching, and native Windows and Linux builds.

The fastest way to run Qwen3.8 27B on Strix Halo (gfx1151)

Source-available C++/CUDA inference engine for Qwen3.8-27B INT4 on 3x RTX 3090