| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| reef_infra-0.1.1-py3-none-any.whl | 2026-09-25 | 1.2 MB | |
| reef_infra-0.1.1.tar.gz | 2026-09-25 | 8.0 MB | |
| README.md | 2026-09-25 | 2.6 kB | |
| v0.1.1 source code.tar.gz | 2026-09-25 | 7.9 MB | |
| v0.1.1 source code.zip | 2026-09-25 | 9.0 MB | |
| Totals: 5 Items | 26.1 MB | 0 | |
What's Changed
- [Feat] Add vLLM token-native capture and runtime kind by @ruizhang0101 in https://github.com/Human-Agent-Society/reef/pull/571
- fix(reefine): rename /evolve command to /reefine by @BobbyZhouZijian in https://github.com/Human-Agent-Society/reef/pull/589
- SAO example: batch 128, paper-shaped runs, results and limitations by @Xuan-1998 in https://github.com/Human-Agent-Society/reef/pull/563
- fix(docs): retain homepage statistics when GitHub requests fail by @BobbyZhouZijian in https://github.com/Human-Agent-Society/reef/pull/590
- fix(runtime): place the training coordinator on the trainer's node by @MikeZheng777 in https://github.com/Human-Agent-Society/reef/pull/587
- feat(records): batch import and storage-owned capacity eviction by @BobbyZhouZijian in https://github.com/Human-Agent-Society/reef/pull/588
- fix(harness): the dsh session reader lifts each event's data beside its envelope by @Benjamin-eecs in https://github.com/Human-Agent-Society/reef/pull/594
- feat(inference): add the vLLM connector that stamps tokens with their weight version by @ruizhang0101 in https://github.com/Human-Agent-Society/reef/pull/601
- docs(harness): clarify how to write a harness method by @YihongT in https://github.com/Human-Agent-Society/reef/pull/606
- fix(docs): load pinned local fonts for reliable builds by @BobbyZhouZijian in https://github.com/Human-Agent-Society/reef/pull/609
- feat(scenario): named release components, one trainer per component, composite recipes (#15 phases 1-3) by @BobbyZhouZijian in https://github.com/Human-Agent-Society/reef/pull/537
- perf(artifact): publish a carried component without re-hashing or cloning it by @BobbyZhouZijian in https://github.com/Human-Agent-Society/reef/pull/652
- refactor(train): name the scenario step scenario_step and keep rollout_id for the checkpoint index by @BobbyZhouZijian in https://github.com/Human-Agent-Society/reef/pull/653
- docs(harness): clarify the harness adapter guide by @YihongT in https://github.com/Human-Agent-Society/reef/pull/607
- feat(reefine): /reefine on every adapter with a command surface, and a proposer that writes what each harness takes by @Benjamin-eecs in https://github.com/Human-Agent-Society/reef/pull/596
- Fix harness release installation and large inference requests by @Xuan-1998 in https://github.com/Human-Agent-Society/reef/pull/600
New Contributors
- @YihongT made their first contribution in https://github.com/Human-Agent-Society/reef/pull/606
Full Changelog: https://github.com/Human-Agent-Society/reef/compare/v0.1.0...v0.1.1
