Skip to content

Actions: NVIDIA-NeMo/RL

Actions

Community Bot

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
community-bot.yml
2,881 workflow runs
2,881 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

feat: adding wandb table log feature, showing concrete test samples Community Bot #2832: Issue comment #2018 (comment) edited by coderabbitai bot
13s
13s
feat: adding wandb table log feature, showing concrete test samples Community Bot #2831: Issue comment #2018 (comment) created by coderabbitai bot
12s
12s
fix: device mismatch when DPO validation at start with CPU offload(Nemotron) Community Bot #2830: Issue comment #1930 (comment) edited by coderabbitai bot
12s
12s
fix: device mismatch when DPO validation at start with CPU offload(Nemotron) Community Bot #2829: Issue comment #1930 (comment) created by coderabbitai bot
13s
13s
fix: Re-enable tests/functional/test_converters.sh functional test Community Bot #2828: Issue comment #2005 (comment) edited by coderabbitai bot
11s
11s
fix: Re-enable tests/functional/test_converters.sh functional test Community Bot #2827: Issue comment #2005 (comment) created by coderabbitai bot
14s
14s
feat: Omni dataloader for HF models Community Bot #2826: Issue comment #2016 (comment) edited by coderabbitai bot
11s
11s
feat: Omni dataloader for HF models Community Bot #2825: Issue comment #2016 (comment) created by coderabbitai bot
12s
12s
perf: Reduce memory footprint for ChunkedDistribuedLogProb Community Bot #2824: Issue comment #1895 (comment) created by nujoug
14s
14s
fix: use smoothed reward metric for VLM GRPO CLEVR convergence tests Community Bot #2823: Issue comment #2015 (comment) edited by coderabbitai bot
10s
10s
fix: use smoothed reward metric for VLM GRPO CLEVR convergence tests Community Bot #2822: Issue comment #2015 (comment) created by coderabbitai bot
13s
13s
[super-pr] feat: add [GPU_DIAG] memory diagnostics logging during worker init Community Bot #2821: Issue #2013 opened by yfw
14s
14s
[super-pr] penalize toolcall in reasoning when it should be outside + invalid thinking tag Community Bot #2820: Issue #2012 opened by yfw
14s
14s
[super-pr] Pass explicit PG to gym Community Bot #2819: Issue #2011 opened by yfw
11s
11s
[super-pr] fix: add support to configure RL+Gym+vLLM port ranges to avoid port contention failures Community Bot #2818: Issue #2010 opened by yfw
11s
11s
[super-pr] fix missing logger Community Bot #2817: Issue #2009 edited by yfw
15s
15s
[super-pr] fix missing logger Community Bot #2816: Issue #2009 edited by yfw
11s
11s
[super-pr] fix missing logger Community Bot #2815: Issue #2009 opened by yfw
15s
15s
[super-pr] overlap NeMo Gym init with vLLM init Community Bot #2814: Issue #2008 opened by yfw
13s
13s
Support multiple dataloader for async grpo Community Bot #2813: Issue #2006 opened by yuki-97
12s
12s
perf: Update moe_token_dispatcher_type default to alltoall Community Bot #2812: Issue comment #2004 (comment) edited by coderabbitai bot
9s
9s
perf: Update moe_token_dispatcher_type default to alltoall Community Bot #2811: Issue comment #2004 (comment) created by coderabbitai bot
11s
11s
perf: shard concat overhead Community Bot #2810: Issue comment #2002 (comment) edited by coderabbitai bot
11s
11s
perf: shard concat overhead Community Bot #2809: Issue comment #2002 (comment) edited by coderabbitai bot
11s
11s
BatchDataDict chunking overhead from repeated concats Community Bot #2808: Issue #2001 edited by pjo256
10s
10s