| | Retire the 2.71 bpw (v2) build: branch removed, kept only as the historical baseline row | avlp12 | Aug 17, 2026 |
| | Speculative decoding opt-in: 10.0-10.6 tok/s on code (+21%) — first double-digit decode; adaptive accept-gate, drafter weights included | avlp12 |
| | v7 MMA fused-codebook prefill: 2k prefill 42.6->24.8s under TP (beats non-TP), KL-certified | avlp12 |
| | Finalize TP campaign: 8.3 tok/s default config, parity+greedy-identity certification, long-prompt stall fixed via chunked prefill | avlp12 |
| | decode: 8.1 tok/s with attention tensor-parallel (K3_ATTN_TP=1); TP long-prompt caveat under jaccl documented | avlp12 |
| | speed: 6.0 -> 6.6 tok/s (dummy-row elision + GLU fusion, bit-exact; MLX 0.32 re-cert) | avlp12 |
| | serving: 5.6 -> 6.0 tok/s (fused MoE router + SiTU-at-store + shared-experts packing) | avlp12 |
| | card: decode now ~5.4 tok/s (kernel vectorization + lazy collectives) | avlp12 |
| | Update README.md | avlp12 |
| | Add files using upload-large-folder tool | avlp12 |