File size: 3,934 Bytes
1613130 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 | {
"artifact": "opencoti-llamafile-0.10.3-c4-x86_64.llamafile",
"sha256": "2f2606573e31b0e02606b546845294add47f68a95be754246b3a58e36e380645",
"sizeBytes": 4812558983,
"builtAt": "2026-07-16T17:51:48.870Z",
"llamafileVersion": "0.10.3",
"opencotiTag": "c4",
"versionString": "opencoti-0.10.3-c4",
"gitCommit": "7d296cd0b35a8631a6de901c6693f92aaceb1e98",
"arch": "x86_64",
"opencotiPatches": [
"0001-lazy-slot-context-defer.patch",
"0002-lazy-slot-context-grow.patch",
"0003-arg-assertion-bypass.patch",
"0004-cuda-std-cpp17.patch",
"0005-lazy-slot-context-shrink.patch",
"0006-kv-reuse-prefix.patch",
"0007-build-header-deps.patch",
"0010-rest-kv-eviction.patch",
"0020-headinfer-per-head.patch",
"0030-neo-pipeline.patch",
"0031-per-stream-split.patch",
"0032-m2-state-io.patch",
"0033-cpy-tie-blck-align.patch",
"0034-m2-pinned-host.patch",
"0035-concat-q-block-aware.patch",
"0036-pcie-probe-consume.patch",
"0040-iqk-flash-attn.patch",
"0042-fused-moe.patch",
"0070-rolling-kv.patch",
"0071-state-io-cosmocc.patch",
"0072-poly-kv-pool.patch",
"0073-turboquant-kv.patch",
"0075-d512-turbo-vec.patch",
"0078-dca.patch",
"0079-gemma-assistant-mtp.patch",
"0080-mtp-d512-fa.patch",
"0081-tcq-q6-asym-kv.patch",
"0082-cuda-build-parallelism.patch",
"0083-tcq-warp-encode.patch",
"0084-cuda-graphs.patch",
"0085-dca-quant-prefill-inlaunch.patch",
"0086-f16-moe-mmf-prefill.patch",
"0087-dca-yarn-factor.patch",
"0088-sparsev-autopolicy.patch",
"0089-sparse-attn-vslash-foundation.patch",
"0090-rolling-kv-compute-reserve.patch",
"0091-multi-session-graph-nodes.patch",
"0092-cuda-build-ccache.patch",
"0093-mtp-dualctx-fused-nextn.patch",
"0094-tinyblas-small-gemm.patch",
"0095-prefill-checkpoint-align.patch",
"0096-remove-single-ctx-gemma-mtp.patch",
"0097-native-elf-cmake.patch",
"0098-rolling-kv-lse-decode.patch",
"0099-dca-gemma4-wiring.patch",
"0100-dca-single-chunk-defer.patch",
"0101-kv-attn-rot-inherit.patch",
"0102-fattn-bulk-h2d-scratch-ring.patch",
"0103-kv-cpu-fa-probe-guard.patch",
"0104-fattn-q6q4-vec-instance.patch",
"0105-ws1-dca-per-tensor-defer.patch",
"0106-ws1-scalar-k-turbo-v-boot-correctness.patch",
"0107-ws1-turbo-mma-fused-decode.patch",
"0108-ws2-dca-fused-turbo-v-inregister.patch",
"0109-cuda-dso-2gb-fatbin-last.patch",
"0110-ws2-asym-band-nan-guard.patch",
"0111-dca-streaming-tail-bug2144.patch",
"0112-rolling-kv-phasegate-prefill-bug2145.patch",
"0113-rolling-kv-reserve-fixpoint-bug2146.patch",
"0114-hybrid-rolling-kv-residency-bug2142.patch",
"0115-rolling-kv-shift-guard-bug2148.patch",
"0116-kv-auto-tier-hal582-p1.patch",
"0117-rolling-kv-shift-rerope-639.patch",
"0118-kv-auto-tier-tstar-spill.patch",
"0119-p0-dcaon-scalar-inregister.patch",
"0120-p2-mixed-kv.patch",
"0121-p2-auto-tail.patch",
"0122-rys-layer-duplication.patch",
"0123-rys-probe.patch",
"0124-rys-mtp-ctx-guard-bug2171.patch",
"0125-hybrid-window-orderagnostic-bug2172.patch",
"0126-p0-dcaoff-scalar-mma-decode.patch",
"0127-p0-dcaoff-scalar-mma-decode-d256-d512.patch",
"0128-mtp-spec-clone-cheap-mmvf-verify.patch",
"0129-rys-probe-nested-args.patch",
"0130-parallel2-fit-ckpt-fixes.patch",
"0131-rys-probe-task-battery-v2.patch",
"0132-runtime-introspection.patch",
"0133-opencoti-version-string.patch",
"0134-cuda-cross-sbsa.patch",
"0135-poly-kv-pool-hybrid-guard.patch",
"0136-polykv-control-plane-api.patch",
"0137-spec-ctx-dft-null-check.patch"
],
"backends": [
{
"backend": "cuda",
"dsoFilename": "ggml-cuda.so",
"sha256": "5e0342ba9b2ded4847280067dfd6ebc4257e1b83a984881988c528c163b23e4a",
"sizeBytes": 4756838104
}
]
}
|