10 Commits

Include renames

Author	SHA1	Message	Date
Salvatore Sanfilippo	f4d03f6cf9	Record backend QA results and preserve failure diagnostics ... Keep failed DSpark command diagnostics and report snapshot prompt/context settings. Extend release checks for attention boundaries, usable ROCm memory, repeated vision embeddings and bounded speculative replay. Record focused GLM and Flash quality, SSD, batching and performance results, including the cold-prefill cost and remaining long-context API and release-matrix gaps.	2026-09-05 19:42:02 +02:00
Salvatore Sanfilippo	be76012927	Batch confident DSpark seeds on M5 and back off on weak drafts	2026-09-04 18:52:56 -07:00
antirez	147109a3cc	Add GLM 5.3 Flash inference ... Implement GLM 5.3 Q2 and Q4 inference across Metal and CUDA, including tensor parallelism, RDMA, batching, MTP, snapshots, SSD streaming, and GLM 5.2 shared-path improvements. Add quantization and validation tools, official quality fixtures, model downloads, regression coverage, release QA, and user documentation.	2026-08-28 23:26:29 +02:00
antirez	abd4ed3885	cli: expose exact MTP sampling mode	2026-08-23 15:05:13 +02:00
antirez	8d683d6616	dspark: add opportunistic temperature sampling	2026-08-23 15:05:13 +02:00
antirez	769a8ba42d	dspark: support exact stochastic speculative decoding	2026-08-23 15:05:13 +02:00
Salvatore Sanfilippo	0e89a0eeff	dspark: commit accepted verifier state directly	2026-08-07 13:01:51 +02:00
antirez	ae504c963e	Update Flash downloads and DSpark support for 0731	2026-08-04 13:00:44 +02:00
antirez	66cbce5289	Unify distributed inference CLI and documentation	2026-07-18 18:50:24 +02:00
antirez	fc9efd1cc7	Add DSpark speculative decoding	2026-07-18 18:47:26 +02:00