antirez b3afc8e991 Document dual-Spark inference and its release checks ...

Describe Q2/MXFP4 network tensor parallelism, DSpark defaults, matching
peer settings, and measured performance with workload and timing scope.
Document native drafter conversion without changing the shipped default.

Extend future QA with physical RDMA, memory ownership, active-stream and
graph checks, official continuation scoring, context boundaries, server
coding sessions, cancellation and fallback coverage.

2026-10-05 14:17:41 +02:00

..

CLIENTS.md

Split README into platform setup and focused reference guides

2026-09-05 14:29:52 +02:00

CUDA_MULTI_GPU.md

Put common getting-started instructions in the README

2026-09-05 19:38:18 +02:00

DGX_SPARK.md

Document dual-Spark inference and its release checks

2026-10-05 14:17:41 +02:00

DISTRIBUTED.md

Document dual-Spark inference and its release checks

2026-10-05 14:17:41 +02:00

METAL.md

DeepSeek v4.1 Flash support for Metal

2026-09-12 11:47:54 +02:00

MODELS.md

Add optimized CUDA inference for Qwen3.8 Flash Next

2026-09-15 15:00:11 +02:00

PERFORMANCE.md

Split README into platform setup and focused reference guides

2026-09-05 14:29:52 +02:00

QWEN38_FLASH_NEXT.md

Document Qwen Metal batching and its quality checks

2026-09-16 15:17:26 +02:00

SERVER.md

Document Qwen Metal batching and its quality checks

2026-09-16 15:17:26 +02:00

SPECULATIVE_DECODING.md

Document dual-Spark inference and its release checks

2026-10-05 14:17:41 +02:00

SSD_STREAMING.md

Document full GLM SSD append gains and validated coding QA

2026-09-06 17:29:02 +02:00

STRIX_HALO.md

Put common getting-started instructions in the README

2026-09-05 19:38:18 +02:00

TESTING.md

Split README into platform setup and focused reference guides

2026-09-05 14:29:52 +02:00