Mirko Covizzi

@MirkoCovizzi · User

GitHub profile ↗ · Compare

[email protected]

21 followers111 repositories

Repositories

MirkoCovizzi/qwen38-27b-rtx5090-mobile

Qwen3.8-27B on a single RTX 3090 with vLLM: ~1,000 tok/s at 64 concurrent (int8 tensor-core GEMMs, fp16 DeltaNet state), ~114 tok/s single-user at default sampling / ~124 greedy (MTP drafts, own-output draft vocab, calibrated int4 lm_head, split-KV verify attention), 150k-262k context; patches, requant scripts, benchmarks

★ 0Forks 0

MirkoCovizzi/ds4

DeepSeek 4 Flash local inference engine for Metal and CUDA

★ 0Forks 0

MirkoCovizzi/pouch

A non-IP protocol for communication between devices and cloud services

★ 0Forks 0