AReal - Browse /v1.0.4 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	2026-05-07	9.1 kB	0
v1.0.4 source code.tar.gz	2026-05-07	9.9 MB	0
v1.0.4 source code.zip	2026-05-07	10.6 MB	0
Totals: 3 Items		20.5 MB	0

What's Changed

fix: FSDP initialization for set-valued wrap class names by @Wangxiaoxiaoa in https://github.com/inclusionAI/AReaL/pull/1187
chore: move figures into assets/figures/ and add community meeting folder by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1192
feat(engine): lora support for MoE models (single node/ cross node) by @gursimar in https://github.com/inclusionAI/AReaL/pull/1159
fix: handle integer device ids in ray rpc server by @Wangxiaoxiaoa in https://github.com/inclusionAI/AReaL/pull/1199
chore: add @CormickKneey as maintainer and agent_service codeowner by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1201
fix: serialize ray object refs in rpc payloads by @Wangxiaoxiaoa in https://github.com/inclusionAI/AReaL/pull/1198
Fix:fix data service failures in IPv6-only environments with HTTP proxy by @TaoZex in https://github.com/inclusionAI/AReaL/pull/1208
feat(service): add external model API support for inference service by @nuzant in https://github.com/inclusionAI/AReaL/pull/1183
feat: add scaffolding rollout workflow by @WeiHaocheng in https://github.com/inclusionAI/AReaL/pull/1064
feat(api): add unified RejectionSamplingConfig for async training by @guozhihao-224 in https://github.com/inclusionAI/AReaL/pull/1088
feat(engine): Add Karmarkar-Karp algorithm by @TaoZex in https://github.com/inclusionAI/AReaL/pull/1151
fix(infra): move data service seed to worker-level config by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1210
docs(community): add first biweekly meeting record and news entry by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1215
chore: update tencent meeting link by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1219
chore: add new maintainers by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1220
refactor(archon): extract inference bridge backends into sglang/vllm submodules by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1221
gov: add maintainer and refine commit convention by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1227
feat: add awex backend for weight update by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1214
chore(deps): upgrade awex to 0.7.0 by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1228
perf(trainer): reduce redundant offload/onload transitions in training loop by @yxs in https://github.com/inclusionAI/AReaL/pull/1163
feat(eval): support for running eval before training by @gursimar in https://github.com/inclusionAI/AReaL/pull/1232
gov: add new maintainer by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1234
gov: add new maintainer by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1235
feat: CP-local loss and configurable CUDA memory profiling by @yulangz in https://github.com/inclusionAI/AReaL/pull/1223
feat: implement megatron weight update with awex by @ran-yan-hk in https://github.com/inclusionAI/AReaL/pull/1239
feat: add disk-mode weight update flow to gateway by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1237
feat(example): add Terminal Bench training example by @ActuallyEdward in https://github.com/inclusionAI/AReaL/pull/1224
gov: update maintainer note by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1248
feat: Implement Megatron PP and CP with Awex for Weight Update by @ran-yan-hk in https://github.com/inclusionAI/AReaL/pull/1246
feat(trainer): add dpo by @HT-Yuan in https://github.com/inclusionAI/AReaL/pull/1190
chore: add uv lock check in pre-commit hooks by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1259
chore(deps): upgrade mbridge from 0.15.1 to 310e8fb by @Adiactive in https://github.com/inclusionAI/AReaL/pull/1258
fix(docker): move venv out of /AReaL to avoid mount override by @sitabulaixizawaluduo in https://github.com/inclusionAI/AReaL/pull/1251
feat: add configurable setup_timeout for data service gateway by @HT-Yuan in https://github.com/inclusionAI/AReaL/pull/1263
test(infra): regression for single-controller SFT batch partitioning by @pratyush618 in https://github.com/inclusionAI/AReaL/pull/1255
fix(engine): eagerly init HCCL subgroups to fix ref compute_logp on NPU by @pratyush618 in https://github.com/inclusionAI/AReaL/pull/1254
feat(experimental): add MegatronEngine awex weight update adapter with EP by @ran-yan-hk in https://github.com/inclusionAI/AReaL/pull/1252
refactor(experimental): rename WeightUpdateAdapter to AwexAdapter by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1269
fix: teardown tcpstore race by @HT-Yuan in https://github.com/inclusionAI/AReaL/pull/1244
refactor(experimental): reuse HTTP clients, add response models, and parallelize ops by @guozhihao-224 in https://github.com/inclusionAI/AReaL/pull/1253
fix(engine): avoid duplicating multimodal tensors by @Wangxiaoxiaoa in https://github.com/inclusionAI/AReaL/pull/1272
chore(experimental): suppress HTTP service logging to warning level by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1274
fix: Update CI test GCP Image by @sitabulaixizawaluduo in https://github.com/inclusionAI/AReaL/pull/1277
feat(experimental): add offload/onload endpoints for inference service by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1276
feat(infra): add n_gpus_per_node abstract property to Scheduler API by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1275
feat(engine): support direct engine construction via from_pretrained without config dataclass by @chenzhiyi021 in https://github.com/inclusionAI/AReaL/pull/1140
feat: introduce Daytona cloud sandbox backend by @mu-hashmi in https://github.com/inclusionAI/AReaL/pull/1231
fix(archon): support Qwen3.5 text checkpoints by @rchardx in https://github.com/inclusionAI/AReaL/pull/1278
chore: update agenda for the next biweekly sync by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1285
fix: apply_chat_template compatibility with transformers>=5.0 by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1280
feat(engine): add Megatron support for Qwen2.5-VL by @Adiactive in https://github.com/inclusionAI/AReaL/pull/1281
refactor(service): rename service controllers and unify service controller configs by @nuzant in https://github.com/inclusionAI/AReaL/pull/1265
chore(dev): add /upgrade-deps skill by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1288
fix(megatron): pass absolute total_train_steps as lr_decay_steps by @yulangz in https://github.com/inclusionAI/AReaL/pull/1289
fix(megatron): compensate pipeline schedule's /num_microbatches in grad by @fishcrap in https://github.com/inclusionAI/AReaL/pull/1273
perf(engine): avoid duplicating multimodal tensors in MegatronEngine by @Adiactive in https://github.com/inclusionAI/AReaL/pull/1291
fix(infra): drain RTensor _fetch_buffer on all consumer workers by @guozhihao-224 in https://github.com/inclusionAI/AReaL/pull/1282
fix(infra): gate RTensor.remotize on DP-head to plug _storage leak by @Adiactive in https://github.com/inclusionAI/AReaL/pull/1297
feat(engine): add Qwen3-VL dense and MoE support to Megatron path by @Adiactive in https://github.com/inclusionAI/AReaL/pull/1301
refactor(inference-service): remove unused release_capacity endpoint by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1293
Add mbridge weight save as an option in Megatron by @hlyli in https://github.com/inclusionAI/AReaL/pull/1300
chore(deps): upgrade runtime dependencies and CI workflow by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1206
chore: upgrade GCP CI image by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1303
fix(sft): report CP-invariant token-count stats (#1242) by @yulangz in https://github.com/inclusionAI/AReaL/pull/1249
gov: add 0501 community meeting materials by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1306
gov: enforce 2-approval merge policy on main by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1307
chore: bump v1.0.4 by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1311