Alpamys 6ec36b5ff6 feat: live FSDP-shard consolidation + serve KV-cache type + ONNX QA (v0.71.14) ... Close the doable tail of the export-QA + deferred-stub family. - #96 consolidate_shards: lazy-torch safetensors merge, TOCTOU-hardened (enforce_under_cwd_and_no_symlink, weights_only=True, atomic_write_bytes), 16 GiB/shard cap, dup-key shape-conflict reject. --plan-only flag. - #140 apply_kv_cache_type -> KvCacheRuntime; soup serve --kv-cache-type (bf16/f16 dtype, q8_0 quantized cache + hqq probe, fp8 Hopper-gated). - #71 ONNX export QA: tiny-GPT2 PASS, TinyLlama host-RAM-bound (qa doc). - #70/#72/#144/#74/#79 deferred to INFRA-BLOCKED tail (kept open). Tests 13430 -> 13476 (+46). ruff clean. Suite green, 78.52% cov.		2026-06-05 13:39:36 +05:00
..
v053_qa.md	feat(v0.53.7): Data Forge + Pipeline live (wave 1) — 11 items	2026-05-13 19:28:11 +05:00
v0716_qa.md	feat(v0.71.6): live build runner + Magpie generator + 2PL/3PL IRT + augment fix	2026-06-02 22:06:12 +05:00
v07114_qa.md	feat: live FSDP-shard consolidation + serve KV-cache type + ONNX QA (v0.71.14)	2026-06-05 13:39:36 +05:00