AReal - Browse /v1.0.2 at SourceForge.net

The interactive file manager requires Javascript. Please enable it or use sftp or scp.
You may still browse the files here.

Name	Modified	Size	InfoDownloads / Week
Parent folder
README.md	2026-03-17	6.3 kB	0
v1.0.2 source code.tar.gz	2026-03-17	8.8 MB	0
v1.0.2 source code.zip	2026-03-17	9.2 MB	0
Totals: 3 Items		18.0 MB	0

Release Note

A massive thank you to our newest contributors who joined us for this release! The strength of this project lies in the collective expertise of the open-source community, and your work is what moves us forward.

🚀 Model & Architecture Updates

Qwen3.5 Support: Added support for both dense and MoE (Mixture-of-Experts) variants of Qwen3.5 (archon backend, DP-only).
On-Policy Distillation: Introduced native support for on-policy distillation.
Added opt-in support for Hugging Face kernels and per-layer optimizer steps with a streaming H2D/D2H pipeline for FSDP.

🛠 Infrastructure & Scalability

Docker & Runtime: Split Docker images into specialized sglang and vllm variants to allow different torch verions and fast docker image update.

📖 Documentation & Localization

Bilingual Support: Launched comprehensive bilingual (EN/ZH) documentation, including a new translate-doc-zh command and fixed LaTeX rendering.
New Guides: Added an online proxy mode training guide and revised existing online RL training tutorials.

What's Changed

feat(models): shard vision encoder across Ulysses SP ranks by @aoshen524 in https://github.com/inclusionAI/AReaL/pull/929
docs: fix broken documentation links by adding /en/ language prefix by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/986
[Feat] Add on-policy distillation support by @HwVanICI in https://github.com/inclusionAI/AReaL/pull/964
fix: keep PPO token stats consistent under context parallelism by @yash27-lab in https://github.com/inclusionAI/AReaL/pull/990
Upgrade GitHub Actions for Node 24 compatibility by @salmanmkc in https://github.com/inclusionAI/AReaL/pull/993
Upgrade GitHub Actions to latest versions by @salmanmkc in https://github.com/inclusionAI/AReaL/pull/994
chore: add Python 3.11 support (requires-python >=3.11,<3.13) by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/991
chore(ci): replace format-check with pre-commit CI and add commit-msg hooks by @rchardx in https://github.com/inclusionAI/AReaL/pull/998
feat(docs): add bilingual documentation with translate-doc-zh command by @ZiyiTsang in https://github.com/inclusionAI/AReaL/pull/995
refactor: flatten sub-module imports to use parent package re-exports by @NJX-njx in https://github.com/inclusionAI/AReaL/pull/996
feat(infra): split Docker image into sglang and vllm variants by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/985
docs: fix broken LaTeX rendering in bilingual docs by @rchardx in https://github.com/inclusionAI/AReaL/pull/1004
fix: CPU-only support on macOS by @zhanghaotong in https://github.com/inclusionAI/AReaL/pull/1003
feat(archon): add moe_router_dtype config for FP32 router gate GEMM by @rchardx in https://github.com/inclusionAI/AReaL/pull/1009
refactor: replace string literals with enums and fix logging issues- … by @HT-Yuan in https://github.com/inclusionAI/AReaL/pull/1008
feat(fsdp): add per-layer optimizer step with streaming H2D/D2H pipeline by @aoshen524 in https://github.com/inclusionAI/AReaL/pull/983
Update NPU doc for v1.0.1 release by @HwVanICI in https://github.com/inclusionAI/AReaL/pull/1022
feat(archon): add Qwen3.5 dense and MoE model support (DP-only) by @rchardx in https://github.com/inclusionAI/AReaL/pull/1012
fix: LoRA and XCCL openai_serving_models LoRA versioning bug by @TinLongYu in https://github.com/inclusionAI/AReaL/pull/1021
fix: unify RPC error response JSON key to "error" across server and s… by @HT-Yuan in https://github.com/inclusionAI/AReaL/pull/1019
fix: close sockets on bind failure, fix exit traceback in trainers by @mango766 in https://github.com/inclusionAI/AReaL/pull/1032
fix(vllm): harden runtime LoRA alias handling for XCCL updates by @Wangxiaoxiaoa in https://github.com/inclusionAI/AReaL/pull/1039
fix(ci): minor fixing ruff format by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1041
docs(github): improve PR template checklist and type-of-change section by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1042
docs: add online proxy mode training guide by @Zijun9 in https://github.com/inclusionAI/AReaL/pull/1006
Add opt-in support for Hugging Face kernels by @lewtun in https://github.com/inclusionAI/AReaL/pull/1033
fix(archon): Wrap router gate in nn.Module for DTensor hook compatibility by @fishcrap in https://github.com/inclusionAI/AReaL/pull/1029
chore(deps): bump sglang, vllm, megatron-core and restructure Dockerfile by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1010
ci(infra): add GCP image baking workflow for CI acceleration by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1045
refactor(infra): simplify RTensor to single-shard and adopt per-trajectory list pipeline by @fishcrap in https://github.com/inclusionAI/AReaL/pull/1017
ci(infra): add GCP image baking workflow and update CI image by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1047
docs: revise online RL training tutorial for EN and ZH by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1049
chore: prepare v1.0.2 release by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1050