Advanced Optimizers (AIO)

A comprehensive, all-in-one collection of state-of-the-art optimization algorithms for deep learning. Designed for maximum efficiency, minimal memory footprint, and superior performance across diverse model architectures and training scenarios.

📦 Installation

pip install adv_optm

Requires PyTorch 2.3+ for torch.compile support.

What's New

🌟 Version 2.5.x: The Massive Refactor

This major update introduces a complete architectural refactor of the library:

🆕 New Optimizers & Scaling

SinkSGD_adv: Added a powerful new optimizer to the lineup.
Spectral Scaling: Now available across all optimizers, achieving width/rank invariant updates for highly stable training.

💾 Memory & State Precision Control

Granular State Precision (state_precision): Drastically reduce memory overhead with new optimizer state modes:
- factored (Rank-2 factored mode)
- fp32 (Full precision)
- bf16_sr & int8_sr (BF16/Int8 with Stochastic Rounding)
Factored Second Moment (factored_2nd): Available for all Adam variants. Works seamlessly alongside any state_precision setting to further slash memory usage.

⚙️ Advanced Dynamics & Momentum

Variance Normalized Momentum (normed_momentum): Applies optimizer normalization before momentum (Normalization then Momentum/NtM). Available for AdamW_adv, SignSGD_adv, and SinkSGD_adv.
Universal Nesterov Momentum: Replaced the hard-to-tune Simplified_AdEMAMix with Nesterov momentum (nesterov) and a dedicated coefficient (nesterov_coef) across all optimizers.
Preconditioning & Signs:
- Added Variance/Confidence Preconditioning (snr_cond) for SignSGD_adv and SinkSGD_adv (requires normed_momentum). Read the technical reports: AASS & sink-v.
- Added Adaptive Stochastic Sign with $L_\infty$ preconditioning (stochastic_sign) for SignSGD_Adv and Lion_adv.
Improved CANS (accelerated_ns): Enhanced for Muon variants by integrating a dynamic lower bound.
New OrthoGrad modes (orthogonal_gradient): Standard OrthoGrad flattened and a new matrix-wise mode iterative.

⚓ Weight Decay Innovations

Centered Weight Decay (centered_wd): Pulls weights toward their pre-train state (anchor). To save memory, anchor precision (centered_wd_mode) can be set to full, float8, int8, or int4.
Fisher Weight Decay (fisher_wd): Now available for Adam variants based on the FAdam paper.
Geometric Weight Decay: Added specifically for SinkSGD_adv and SignSGD_adv.

(Note: Lion_Prodigy_adv, Simplified_AdEMAMix, and heuristic cautious/grams modes have been deprecated in favor of these superior, theoretically-grounded features).

Click to see older release notes (v1.2.x - v2.1.x)

Version 2.1.x

New Optimizer: Added Signum (SignSGD with momentum) to the SignSGD_adv family.

Version 2.0.x

⚡ torch.compile Support: Fully implemented for all advanced optimizers. Enable via compiled_optimizer=True to heavily fuse and optimize the optimizer step path.
📉 1-Bit Factored Mode: Vastly improved implementation via nnmf_factor=True.
🛠️ Broad performance and stability improvements across all optimizers.

Version 1.2.x

Advanced Muon Variants: Brought the groundbreaking Muon optimizer into the fold, enriched with features from recent literature.

Optimizer	Description
`Muon_adv`	Advanced Muon implementation featuring CANS, NorMuon, Low-Rank Orthogonalization, and more.
`AdaMuon_adv`	Combines Muon's geometry with Adam-like adaptive scaling and sign-based orthogonalization.

Prodigy Speedup: Prodigy variants are now 50% faster by eliminating unnecessary CUDA syncs (Shoutout to @dxqb!).
Stochastic Rounding for BF16: Parameter updates and weight decay now accumulate in float32 and round once at the end.
Cautious Weight Decay: Implemented for all advanced optimizers (Paper).
Fused Operations: Transitioned to fused and in-place operations wherever possible.

💡 Core Innovations

(Documentation expanding on the theory and usage of these features is coming soon!)

Release files for adv-optm 2.6.2.dev1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for adv-optm 2.6.2.dev1
File	Size	Uploaded
adv_optm-2.6.2.dev1.tar.gz	61.9 kB	Sep 24, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for adv-optm 2.6.2.dev1
File	Interpreter	ABI	Platform	Reset
adv_optm-2.6.2.dev1-py3-none-any.whl 87.1 kB Sep 24, 2026	Python 3	none	any	Details

Total release size: 149.0 kB

Release files / adv_optm-2.6.2.dev1.tar.gz

Download URL	adv_optm-2.6.2.dev1.tar.gz
Size	61.9 kB
Tags	Source
SHA-256 checksum How to use checksums	`5d96ba5b0bbc19c59ce032584054b51ed8b64a985e9b49b1a25c6bcc04796fa8`
BLAKE2b-256 checksum How to use checksums	`45af67ffdda0c2cca7bf2592928d24153e468d3d735013e58b156c953b86fdcd`
Upload date	Sep 24, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/6.2.0 CPython/3.12.10`

Release files / adv_optm-2.6.2.dev1-py3-none-any.whl

Download URL	adv_optm-2.6.2.dev1-py3-none-any.whl
Size	87.1 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`8f662bda650e2e1ec7e7a4d6a014f31268b4a9d9220a1b9dcf2c2222bbb7368c`
BLAKE2b-256 checksum How to use checksums	`5e4ae17dd7210e160aab92428ceec119cac48fffffddf062ae3d6adf790d5e35`
Upload date	Sep 24, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/6.2.0 CPython/3.12.10`