v2.5.0

What's Changed

• Rename scatter_object to broadcast_object() by @epwalsh in #471
• Add exponential LR scheduler by @baileykuehl in #472
• Fix typos in code by @agupta2304 in #473
• Add Peri-LN transformer block, module for residual stream by @epwalsh in #457
• Fix QK norm parameter count in AttentionConfig.num_params() by @dirkgr in #477
• Update docker images and versions (most recent image - torch 2.9.1, cuda 13.0) by @tyler-romero in #459
• Internal olmo3 32b midtrain/long context scripts by @tyler-romero in #450
• Use correct peak LR in Olmo 3 32B midtrain by @soldni in #481
• Load multiple checkpoints at the same time and average them by @dirkgr in #396
• Fix type annotation issue in NumpyInterleavedFSLDataset (Issue #486) by @ada-ggf25 in #487
• Composable data loading by @epwalsh in #416
• [MoE] Upstream TrainModuleConfig by @tyler-romero in #490
• various improvements to set up for new model ladder by @epwalsh in #493
• set num_execution_units default to 1 by @epwalsh in #495
• split off transformer and attention CPU tests by @epwalsh in #496
• Changes the prefix used by GPUMemoryMonitorCallback to avoid using the reserved system prefix by @finbarrtimbers in #497
• Avoid torch recompile when intra-doc masking enabled by @epwalsh in #504
• Gated Attention by @tyler-romero in #508
• Model ladder, revamped by @epwalsh in #494
• No-global-rope (GNoPE) by @tyler-romero in #509
• Fix metrics filter by changing system/ prefix to new gpu_memory/ prefix by @finbarrtimbers in #499
• Add in-loop LM evals to ladder by @epwalsh in #512
• Bottom-up flops estimation & total flops tracking by @tyler-romero in #511
• MXFP8 Linear Support by @tyler-romero in #510
• ladder lmevaluator fix by @tyler-romero in #514
• Fix overflow when too many global flops are computed by @tyler-romero in #515
• Jacobm final merge sft by @jacob-morrison in #400
• Add instance filter to olmo3 ladder by @tyler-romero in #516
• Make data loading preprocessing more robust to race conditions by @epwalsh in #517
• Add scripts for slurm env by @epwalsh in #518
• update slurm scripts by @epwalsh in #519
• Allow cordoning nodes by @epwalsh in #520
• More slurm improvements by @epwalsh in #521
• Add temp & mem healthchecks and allow comments in cordoned nodelist by @tyler-romero in #522
• Consolidate bash helper functions by @epwalsh in #523
• Increase ulimit -n within slurm jobs by @tyler-romero in #524
• More Slurm util improvements by @epwalsh in #525
• Ladder improvements by @epwalsh in #529
• GAP Monitor Improvements by @tyler-romero in #527
• Ulysses CP Support by @tyler-romero in #532
• Support for Muon and Dion Optimizers by @tyler-romero in #528
• Add a model ladder for "more norms" and other internal ladder improvements by @epwalsh in #535
• Add 60M and 1M param model configs by @tyler-romero in #536
• Add Qwen3 model support (0.6B-32B) by @finbarrtimbers in #533
• update to Beaker-py v2, use gantry API internally by @epwalsh in #540
• Add 60M and 100M param ladder rungs by @tyler-romero in #543
• always include GOOGLE_CREDENTIALS by @epwalsh in #545
• Ensure final metrics always logged by @epwalsh in #546
• Log chinchilla multiple from SpeedMonitorCallback by @epwalsh in #548
• improve beaker client caching by @epwalsh in #549
• Add a stepped WSDS variant, integrated into ladder by @epwalsh in #538
• update base LR for stepped schedule by @epwalsh in #551
• Add Gemma 3 support with HF weight conversion by @finbarrtimbers in #534
• Add 14M param config by @tyler-romero in #552
• Fix flaky and broken transformer model tests by @tyler-romero in #537
• Add model ladders for recently run ablations by @tyler-romero in #553
• Fix 1M model by @tyler-romero in #555
• Now, WandBCallback and CometCallback log the Beaker URL to their config sections. by @finbarrtimbers in #547
• Skip init_process_group if already initialized by @finbarrtimbers in #539
• Fix failing test_build_world_mesh_cpu for pytorch 2.10 by @tyler-romero in #559
• Easier discovery and collection of ladder results by @tyler-romero in #556
• Reduce disk usage of convert_checkpoint_to_hf_test by @tyler-romero in #561
• More ladder work by @epwalsh in #560
• Add torch 2.10 image, fix FA3/TE incompatibility by @epwalsh in #562
• Add support for FA4 (CUTE implementation) by @epwalsh in #563
• Use uv in CI by @epwalsh in #564
• Add Callback.pre_log_metrics() method by @epwalsh in #566
• Replace OmegaConf with dataclass-extensions by @epwalsh in #567
• Utilize Registrable for optim/scheduler configs by @epwalsh in #568
• Integrate QuACK kernels, minor improvements to Docker build by @epwalsh in #569
• SequenceMixer as base for both attention and recurrent layers by @tyler-romero in #570
• update deps by @epwalsh in #574
• Add GatedDeltaNet by @tyler-romero in #572
• Add option to load downstream eval tasks lazily by @epwalsh in #575
• [QoL] include throughput metrics when updating Beaker workload description by @epwalsh in #577
• Ensure Beaker workloads submitted from CI are canceled gracefully by @epwalsh in #579
• Evaluate upon completion of training by @tyler-romero in #580
• Fix documentation link for OLMo-core Beaker images by @natolambert in #581
• Update SFT documentation with tokenization and training tips by @natolambert in #571
• Join bookkeeping ops before checkpointing by @epwalsh in #583
• Propagate gantry launch options by @tyler-romero in #582
• Fix some logic around soft timeouts in launch.beaker by @epwalsh in #584
• Make SLACK_WEBHOOK_URL not actually required by @tyler-romero in #585
• Optional disk-backed caching for remote IO ops, data loading config improvements by @epwalsh in #586
• Add chat template verification step to SFT README by @natolambert in #590
• Checkpointing improvements by @epwalsh in #588
• GatedDeltaNet improvements by @tyler-romero in #592
• Bump dataclass-extensions to fix pickle issue by @epwalsh in #595
• Fix config encoding/merging with non-init fields by @epwalsh in #596
• Internal cleanup of composable sampling classes by @epwalsh in #599
• Fix config decode when CLASS resolves a Registrable subclass by @baileykuehl in #600
• port over @dirkgr's fix for async callbacks by @epwalsh in #601
• Fix GDN initialization by @tyler-romero in #602
• Mark ephemeral checkpoints in metadata by @epwalsh in #605
• Add claude md for Olmo-core by @tyler-romero in #611
• Support flattening 3d parameters to 2d for use with Muon by @tyler-romero in #606
• docs: SFT conversion tips for tokenizer override and parallel jobs by @natolambert in #608
• updates to claude.md file by @epwalsh in #612
• Support block-pattern based initialization of hybrid transformers by @tyler-romero in #607
• Ladder5 v2 by @dirkgr in #610
• Add model merging callback with integration test by @baileykuehl in #558
• Fix A100 peak flops spec in speed monitor by @tyler-romero in #614
• Add token ID validation guard in DataCollator by @tyler-romero in #619
• fix typo in mixes init by @baileykuehl in #621
• Support perplexity evals with context and tensor parallelism by @tyler-romero in #616
• Add training smoketest skill for claude by @tyler-romero in #618
• Hybrid model pre/mid/lc configs by @tyler-romero in #613
• Add midtraining ingredient to to hybrid checkpoint list by @tyler-romero in #633
• Add new in-loop eval tasks and update eval dependency to PyPI release by @kyleclo in #587
• Full support for flash attention 4 (including generation kv cache support, installation from pypi) by @tyler-romero in #642
• Add documentation for native generation module and chat interface by @tyler-romero in #641
• Add Code Fresh per-language perplexity eval by @kyleclo in #639
• fix attn backend forcing for TransformerGenerationModule by @tyler-romero in #645
• ✨ Quality: Add documentation about checkpoint conversion limitations by @lukebaze in #648

New Contributors

• @agupta2304 made their first contribution in #473
• @ada-ggf25 made their first contribution in #487
• @finbarrtimbers made their first contribution in #497
• @jacob-morrison made their first contribution in #400
• @natolambert made their first contribution in #581
• @kyleclo made their first contribution in #587
• @lukebaze made their first contribution in #648

Full Changelog: v2.4.0...v2.5.0