Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦
C
Commits on main
fix(build): track SSE4.1 kernel header dependencies
a3ec030 codex 20260920-012741-906614622 committed 2 months ago
fix(sse41): preserve grouped-int4 scalar order
e09b3db codex 20260830-005353-2548412838 committed 2 months ago
perf(sse41): vectorize grouped-int4 GEMV
4d35744 codex 20260829-144356-267014324 committed 2 months ago
Merge dev into feat/glm53-assistant-continuation: keep the max_tokens ceiling paragraph next to the continuation one
836310f JustVugg committed 1 month ago
Merge pull request #1692 from JustVugg/release/1.12.1-notes-wave3
b94accc Vincenzo Fornaro committed 1 month ago
Merge pull request #1556 from karlem/fix/macos-metal-gpu-discovery
2cc1040 Vincenzo Fornaro committed 1 month ago
Merge pull request #1580 from LTCjRet/fix/windows-qwen36-cuda-dll
5f1cb69 Vincenzo Fornaro committed 1 month ago
fix(olmoe): wait for an expert already being read instead of reading it twice
2020aec Marcello Costagliola committed 1 month ago
Merge pull request #1597 from kevin9327/fix/serve-tool-call-arguments-non-object
d8c4209 Vincenzo Fornaro committed 1 month ago
Merge branch 'dev' into feat/windows-xdna-shared-expert-lane
2a3b17c Kenneth-Javier Ortigoza committed 1 month ago
test(fp8): e2e serve-batch pin, refusal canary, and mint-to-load regression
ea8a7e4 monotophic committed 2 months ago
feat(cuda): add fmt=8 (fp8-e4m3) decode to the absorb kernels
9c7ae38 monotophic committed 2 months ago
refactor(quant): move FP8_BLOCK/fp8_nblk into fp8_format.h
6826c00 monotophic committed 2 months ago
Merge pull request #1605 from ekalb81/fix/glm-oracle-validation
9e152d4 Vincenzo Fornaro committed 1 month ago
Merge pull request #1622 from Frank-zhu0404/fix/issue-1615-per-matrix-direct-read
4efbb8d Vincenzo Fornaro committed 1 month ago
Merge pull request #1568 from Yoruxyv/feat/indonesian-dashboard-translation
b6955b6 Vincenzo Fornaro committed 1 month ago
CHANGELOG 1.12.1: the third wave of contributor pull requests
9e85af7 JustVugg committed 1 month ago
Merge pull request #1618 from bherald/docs/qwen38-vision-overview-20260918
0eb3a87 Vincenzo Fornaro committed 1 month ago
Merge pull request #1624 from namespaceMarcello/fix/build-warnings
11106c9 Vincenzo Fornaro committed 1 month ago
Merge dev into fix/macos-metal-gpu-discovery: keep the int8 trunk line in the VRAM branch
232b282 JustVugg committed 1 month ago
Merge dev into fix/windows-qwen36-cuda-dll: keep dev's qwen36 prerequisites plus the alias and .build-config
9fc4ca3 JustVugg committed 1 month ago
Merge pull request #1647 from Suraj2105-1/fix/node20-eol-release-pipeline
96ef114 Vincenzo Fornaro committed 1 month ago
Merge dev into fix/serve-tool-call-arguments-non-object: keep both tool-call regressions
5480ea5 JustVugg committed 1 month ago
Merge pull request #1649 from Suraj2105-1/fix/alpine-3.21-eol-upgrade-to-3.24
ac94bb0 Vincenzo Fornaro committed 1 month ago
Merge pull request #1651 from Suraj2105-1/fix/pyflakes-code-bugs
9e2555a Vincenzo Fornaro committed 1 month ago
Merge pull request #1559 from kreuzzelg/qwen36-down-bits
2051c18 Vincenzo Fornaro committed 1 month ago
Merge pull request #1690 from JustVugg/release/1.12.1-notes-batch
9ddd00a Vincenzo Fornaro committed 1 month ago