Changelog

✨ New Features

• SSE history can now be extracted and manipulated in Aleph allowing for user-implemented estimators.
• Helper function, sse_order to determine the order from a given SSE configuration.
• Added suite of fermionic operator transformation functions to simplify and manipulate fermionic operators:
  • fermionic_site_reorder(operator): Reorders creation and annihilation operators by site while tracking phase changes from anticommutation relations.
  • converted_to_create_destroy(operator): Converts named fermionic operators to their creation/annihilation form. Handles multi-site interactions like density-density ($n_i n_j = c_i^\dagger c_i c_j^\dagger c_j$), hopping ($c_i^\dagger c_j + c_j^\dagger c_i$), and pair operators ($c_i^\dagger c_j^\dagger c_k c_l$).
  • simplified_create_destroy(operator): Simplifies operator products by reordering and removing invalid terms (e.g., $c_i c_i = 0$ and $c_i^\dagger c_i^\dagger = 0$).
  • fused_into_number(operator): Fuses creation-annihilation pairs into number operators ($c_i^\dagger c_i \to n_i$).
  • reduced_number_power(operator): Reduces powers of number operators to at most 1 ($n_i^2 = n_i$ for fermionic systems).

🐛 Fixes

• Fixed SSE specific heat capacity estimator.

⚙️ Internal / Technical Changes

• Slight speed improvement to SSE sampling.

Release description not available.

🚀 Highlights

• Kai updated to leverage the latest Aleph APIs for faster and more efficient responses

🛠 Improvements

• Kai system prompt optimization
  • Updated to take advantage of the latest Aleph APIs.
  • Improved efficiency and response speed.

🚀 Highlights

• Patch release improving cloud job reliability

🐛 Fixes

• Fixed a race condition that could prevent some cloud jobs from starting correctly.

🚀 Highlights

• Fix for a Kai error affecting certain prompts
• Runtime upgrade for improved performance and security

🐛 Fixes

• Fixed an issue in Kai that could trigger a "finishReason" error for certain prompts.

🛠 Improvements

• Runtime upgrade
  • Improved overall performance.
  • Enhanced security and stability.

🚀 Highlights

• Figures are now editable directly in Workshop after creation
• Smarter file renaming behavior
• New “Format Document” action with shortcut support
• Major improvements to Markdown editing and navigation

✨ New Features

• Editable figures

  • Figures can now be edited directly in Workshop after they are generated.
  • Makes iterative visualization workflows smoother and faster.

• Format Document action

  • New Format Document action available via:
    • Shortcut: Alt/Opt + Shift + F
    • Toolbar button (shown when a formatting provider is available)
  • Provides consistent formatting across supported file types.

• Enhanced Markdown navigation

  • Clicking links in Markdown preview now navigates to other workspace files.
  • Markdown-to-Markdown links open directly in preview mode for a smoother reading experience.

🛠 Improvements

• File renaming UX

  • Only the filename (not extension) is selected by default.
  • Warning shown when changing file extensions to prevent accidental type changes.

• Markdown editing upgrades

  • Hover previews
  • Automatic link detection
  • Block folding
  • General editing polish and usability improvements

🐛 Fixes

• Minor UI and interaction fixes across editing workflows.

🚀 Highlights

• Expanded the library of symbolic Fermionic operators in Aleph with new multi-site and parameterized interactions. See the New Features section for details.
• Forge scripts can now declare arguments.

✨ New Features

• Significant expansion of symbolic Fermionic operators in Aleph, adding standardized factory functions for several interactions:
  • NumberNumber(i, j): Two-site density-density interaction $n_{i} n_{j} = f_{i}^\dagger f_{i} f_{j}^\dagger f_{j}$.
  • TransferNumber(i, j, k): $f_{i}^\dagger f_{j} n_{k}$ interaction.
  • HopNumber(i, j, k): $(f_{i}^\dagger f_{j} + f_{j}^\dagger f_{i}) n_{k}$ interaction.
  • NumberSum(sites): Total occupancy on a set of sites $\sum_{i \in \mathrm{sites}} n_{i}$.
  • NearestNumberSum(sites): Nearest-neighbor density interaction $\sum_{i} n_{i} n_{i+1}$.
  • NextNearestNumberSum(sites): Next-nearest-neighbor density interaction $\sum_{i} n_{i} n_{i+2}$.
  • PairTransfer(i, j, k, l): Directed pair transfer operator $f_{i}^\dagger f_{j}^\dagger f_{k} f_{l}$.
  • PairHop(i, j, k, l): Pair hopping operator $f_{i}^\dagger f_{j}^\dagger f_{k} f_{l} + \mathrm{h.c.}$.
  • PhaseHop(angle, i, j): Complex phase-hop operator $e^{i \phi} f_{i}^\dagger f_{j} + e^{-i \phi} f_{j}^\dagger f_{i}$.
  • NumberPhase(angle, i): Site-specific phase operator $\exp(i \phi n_i)$.
  • NumberNumberPhase(angle, i, j): Two-site phase operator $\exp(i \phi n_{i} n_{j})$.
  • ModeRotation(angle, i, j): Givens rotation operator $\exp(i \theta (f_{i}^\dagger f_{j} + f_{j}^\dagger f_{i}))$.
  • PairRotation(angle, i, j): Bogoliubov rotation operator $\exp(i \theta (f_{i}^\dagger f_{j}^\dagger + f_{j} f_{i}))$.
  • NumberPotential(coefficients [, sites]): Efficient representation of weighted density sums $\sum_{i \in \mathrm{sites}} \alpha_i n_i$.
  • CoulombSum(tensor): General Coulomb interaction $\sum_{i,j,k,l,\sigma,\sigma'} V_{ijkl} a_{i\sigma}^\dagger a_{j\sigma'}^\dagger a_{k\sigma'} a_{l\sigma}$ on interleaved spin-orbitals.
• Unified Fermionic operator naming conventions and added legacy aliases (FourBody, DoubleExcitation, NumberN, NumberNumberNN, NumberNumberNNN) for backward compatibility.
• Added Discrete Cosine Transform (DCT) and Discrete Sine Transform (DST) functions with support for types I-IV, custom normalization, and orthonormal scaling. Both transforms support 1D and 2D operations on Matrix and Array types.
• Forge scripts can now declare arguments and retrieve them with get_arg(name, type[, default]), enabling typed inputs to be provided when launching executions or scheduling jobs.
• Added function to compute the specific heat from the order estimator sse_specific_heat(order).
• The history of an SSE simulation can now be recorded and read/written from/to an HDF5 file allowing for checkpointing of simulations and for post processing.

🐛 Fixes

• Added string parsing support for complex conversion in Aleph, so expressions like complex("2 + 1i") and "2 + 1i".complex() now work.
• Fixed LSP semantic token locations for interpolated member access (for example ${ab.first}), avoiding invalid and overlapping semantic tokens in editors.
• Mixed-complexity operator expressions that previously caused errors now work correctly. For example, 2.0 * X(0) + 3.0i * Y(1) is now a valid expression.
• Fbit and Qbit can faithfully represent BasisStates with up to 64 sites, and will throw an error if an attempt is made to create a BasisState with more than 64 sites. This prevents silent truncation of states that was previously occurring.

⚠️ Breaking Changes

• Failed string-to-number conversions now throw an error instead of defaulting to 0. This affects conversions like real("abc") and integer("abc"). Loss-of-precision conversions still behave as before, e.g. integer("5.3") -> 5.

🚀 Highlights

• Expanded figure editing capabilities (limited preview)
• Minor bug fixes and stability improvements

✨ New Features

• Figure editing (limited preview)
  • Additional editing options are now available when working with figures.
  • Continued enhancements as part of the limited preview experience.

🐛 Fixes

• Minor bug fixes to improve overall reliability and usability.

🚀 Highlights

• Limited preview features introduced (details coming soon)
• Improved error reporting and minor bug fixes

✨ New Features

• Limited preview
  • Early groundwork for upcoming features.
  • More details will be announced in a future update.

🐛 Fixes

• Improved error reporting for a clearer and more consistent experience.
• Minor bug fixes across the Workshop.

🚀 Highlights

• Patch release focused on stability and polish

🐛 Fixes

• Fixed a number of non-critical bugs to improve overall reliability and user experience.

🚀 Highlights

• Patch release addressing Kai credit usage issues

🐛 Fixes

• Updated Kai configuration to enforce a specific AI provider, resolving credit-related issues experienced by some users.

🚀 Highlights

• Stability release with groundwork for upcoming Aleph features
• Fixed dropdown scrolling issue affecting some Firefox users on Linux

🛠 Improvements

• Internal updates preparing Workshop for upcoming Aleph capabilities.

🐛 Fixes

• Fixed a scrolling bug that could impact dropdown menus on Firefox (Linux).

🚀 Highlights

• Patch release focused on monitoring improvements

🚀 Highlights

• Security and error-handling improvements across Workshop
• File move cancellations no longer trigger error messages

🛠 Improvements

• Security & reliability
  • General improvements to security and error-handling behavior for a more stable experience.

🐛 Fixes

• Fixed an issue where aborting a file move operation could incorrectly display error messages.

🚀 Highlights

• Improved handling and visibility of transient errors
• Internal groundwork for upcoming bindings

🛠 Improvements

• Transient error handling

  • More robust detection and reporting of temporary failures to improve reliability and troubleshooting.

• Platform readiness

  • Background updates preparing the platform for upcoming bindings and integrations.

✨ New Features

• Added Figure.visual_map(...) options in the Plotting module.
• Automatically add a default visual_map when none specified when a figure has a heatmap.
• MPS and MPO operator "+" perform the exact addition, with output bond dimension equal to the sum of that of each tensor network.
  • Repeated addition cause a linear increase in bond dimension.
  • use .compress to control the growth for MPS.
• MPS and MPO operator "*" perform the exact operator-state and operator-operator mulitplication. Output bond dimension is the product of that of each tensor network.
  • Repeated multiplication causes an exponential increase in bond dimension.
  • use .compress to control the growth for MPS.
• explicit management of boundary condition for MPS and MPO. Important to set it correctly for full compression.
  • .open_boundary() to open the boundary condition, .periodic_boundary() for periodic boundary condition.
  • .set_boundary_like(mps | mpo) to set the boundary condition like another TN.
  • DMRG only works with open boundary condition MPS and MPO. 

🐛 Fixes

• Fixed LSP semantic tokenization of interpolated strings where the implicit string(...) call could mark the start of a literal as a function.
• Added correct semantic token locations inside ${...} interpolation spans.

🚀 Highlights

• Improved API documentation and SDKs
• runtimeId is now optional when creating executions and jobs

🛠 Improvements

• API & SDK enhancements

  • Clearer API documentation.
  • Updated SDKs for better developer experience and consistency.

• Simplified execution & job creation

  • runtimeId is now optional when creating executions or jobs.
  • Defaults automatically to the current Aleph version if not specified.

🧪 Migration Notes

• Existing integrations can omit runtimeId unless a specific runtime version is required.

🚀 Highlights

• Internal pre-release focused on improving interoperability and integration scenarios
• No user-facing changes in this release.

🛠 Improvements

• Backend enhancements to support broader integration use cases.
• Internal changes preparing the platform for upcoming interoperability features.

🚀 Highlights

• Reload individual tabs without refreshing the entire Workshop
• Clickable stack traces in Aleph outputs
• Improved figure rendering, editing, and validation

✨ New Features

• Reload Tab

  • New “Reload Tab” action lets you refresh a single tab if it’s in an error state, without reloading the whole page.

• Clickable stack traces

  • When an Aleph script throws an error, stack traces in the output window are now clickable.
  • Clicking opens the corresponding file in the editor at the correct line and column.

• Figure raw editing enhancements

  • “Format Document” context menu for raw figure files.
  • JSON validation and IntelliSense support when editing figure definitions.

🛠 Improvements

• Figure rendering
  • Fixed aspect ratio in Workshop tabs (3:2), matching export size.
  • Axis ticks and labels are now displayed by default.
  • Visual maps now use the viridis color scale by default.
  • Minor improvements to LaTeX rendering in axis labels.

🐛 Fixes

• Minor UI and rendering fixes across figures and output handling.

🚀 Highlights

• Kai model updated to improve AI code generation quality

🛠 Improvements

• Kai model upgrade
  • Improved code generation accuracy and consistency.
  • Better adherence to project structure and conventions.
  • More reliable diffs and suggested changes.

🚀 Highlights

• Restored LSP diagnostics

🐛 Fixes

• Re-enabled Language Server Protocol (LSP) diagnostics that were accidentally disabled.
• Diagnostics (errors, warnings, hints) are now properly reported again in supported editors and in Kai.

🚀 Highlights

• Smoother file drag’n’drop in the workspace tree
• Easier file and folder creation from more places
• Quick copy actions for Agent ID and Workspace ID
• Execution signals now visible when a process is interrupted

✨ New Features

• Improved drag’n’drop

  • More reliable file drag-and-drop behavior in the file system tree.

• Expanded context menus

  • “New File” and “New Folder” actions are now available directly in more tree locations.
  • Added Copy Agent ID and Copy Workspace ID actions to simplify API integration workflows.

• Execution signal display

  • If a process was interrupted by a signal (e.g., killed or timed out), it is now shown alongside execution details.

🚀 Highlights

• Improved execution diagnostics with signal capture
• Safeguard added to prevent excessive file uploads during executions

✨ New Features

• Execution signal capture
  • In addition to exit codes, the platform now captures the process termination signal.
  • Provides clearer insight into whether a process exited normally or was interrupted (e.g., killed, timeout, etc.).

🛠 Improvements

• Upload safety guard
  • If more than 100 files are created or modified during an execution:
    • All affected files are bundled into a single ZIP archive.
    • The archive is uploaded as-is to the platform.
  • Prevents rogue or misconfigured scripts from uploading excessive numbers of files.

🧪 Migration Notes

• No action required for users.
• Workflows generating large numbers of files may now produce a ZIP artifact instead of individual files.

🚀 Highlights

• Patch release fixing file listing pagination

🐛 Fixes

• Fixed an issue affecting pagination when listing files.

🚀 Highlights

• Rewrite of HDF5File support in Datafiles to enable reading/updating platform-stored files, in addition to writing new files.
• HDF5File now supports reading/writing HDF5 files up to version 2.0 and provides compression options.
• HDF5File supports many Aleph types (lists, arrays, tensors, and more).
• Added TextFile support for reading/writing files as strings.

✨ New Features

• Rewrote HDF5File support in Data Files to allow reading and updating platform-stored files, in addition to writing new files.
• HDF5File now supports reading/writing HDF5 files up to version 2.0 and includes compression options.
• HDF5File supports many Aleph types, including lists, arrays, and tensors.
• Added TextFile for reading/writing files as strings.

⚠️ Breaking Changes

• operator_function API has changed to support both Hermitian and non-Hermitian operators; it now accepts an option map similar to that of iterative_eigensolver. See the New Features section for details.
• HDF5File types and functions have been substantially updated and may break code written prior to this release; this binding is expected to be stable moving forward.

🚀 Highlights

• Kai updated to better support figures and admonitions in generated content

🛠 Improvements

• Kai system prompt update
  • Improved support for generating figures.
  • Improved handling of Markdown admonitions (note, tip, warning, etc.).

🐛 Fixes

• Minor internal improvements to AI content formatting.

🚀 Highlights

• Patch release improving Kai message rendering

🐛 Fixes

• Minor fixes to Kai message rendering to improve formatting and display consistency.

🚀 Highlights

• Job Details page is now fully scrollable
• Markdown preview gains figure JSON support and admonitions
• Figures can now be exported with a light theme
• Long sub-menus are now scrollable

✨ New Features

• Scrollable Job Details

  • The Job Details page is now scrollable, making it easier to navigate long outputs and metrics.

• Figure rendering in Markdown (JSON)

  • Markdown preview now supports figure JSON code blocks and renders them directly.

• Docusaurus-style admonitions

  • Markdown preview now supports admonitions (note, tip, warning, etc.) for richer documentation.

• Light theme export for figures

  • Figures can now be exported with a white background using a light theme.

🛠 Improvements

• Scrollable sub-menus
  • Long sub-menus now scroll properly, improving usability in dense navigation areas.

🐛 Fixes

• Minor UI and layout fixes.

🚀 Highlights

• Internal platform updates and maintenance release

🚀 Highlights

✨ New Features

• Introduced the Plotting module to create figures in Aleph for data analysis, including bar, line, scatter, and heatmap plots.
  • Supports legends, multiple series, multiple plots in the same figure, and a ton of customization options.
• Functions for the computation of matrix elements and expectation values with MPS and MPO.
  • matrix_element(bra,operator,ket) to compute the matrix element $\bra{\phi} O \ket{\psi}$ of an operator.
  • expval(operator,ket) to compute the expected value $\bra{\psi} O\ket{\psi}$ of the operator.
• Added accumulators for binning analysis of time series and markov chain data. Currently support accumulator("no_binning") and accumulator("binning"). The bin size can be set via accumulator("binning" ["bin_size" : 1024 ]).
• Added a FourBody symbolic fermionic operator: $c^{\dag}_{i}c^{\dag}_j c_k c_l$. An operator can be created via FourBody(i,j,k,l) where i,j,k,l are the sites where the operator acts.
• Added a DoubleExcitation fermionic operator: $c^{\dag}_{i}c^{\dag}_j c_k c_l + h.c.$. An operator can be created via DoubleExcitation(i,j,k,l) where i,j,k,l are the sites where the operator acts.
• The array and matrix factories now accept array and matrix objects together with a type tag. For example: array(my_matrix, as_complex) will create a complex array from my_matrix even if it is of real type.
• MPS compression routine. mps.compress(tol) to compress a MPS to specified error tolerance.
• operator_function now supports both Hermitian and non-Hermitian matrices and operators. The appropriate Krylov method (Lanczos or Arnoldi) is automatically selected based on the input. The API can be summarized as:
  • operator_function(H, dim, fun(real){real}, ["is_real":true]) for real-valued functions and Hermitian operators. Note that the ["is_real":true] option is required to indicate that the function is real-valued.
  • operator_function(H, dim, fun(complex){complex}) for other Hermitian operators.
  • operator_function(H, dim, fun(complex, integer){complex}) for non-Hermitian operators. The integer argument stands for the order of the derivative of the function, which is required for non-Hermitian operators.
  • Any of the above can also be called with a matrix Hmat instead of an operator H and a dimension dim.
  • Any of the above also accepts an optional dictionary of options as last argument, e.g. ["krylov_dimension": 32].
• The PhaseHop(angle,i,j) factory function has been added to create fermionic hopping operators with an associated phase factor: $e^{i \cdot angle} c^{\dag}_i c_j + \mathrm{h.c.}$.

🐛 Fixes

• Assigning complex arrays or matrices to real arrays or matrices used to silently fail, it now raises an error.

⚙️ Internal / Technical Changes

• Adds internal support for handling sparse matrix operations with free fermion objects.

• Real arrays and matrices are not implicitly convertible to complex arrays and matrices anymore. An explicit conversion is now required using as_complex.

• Cmake presets to help manage compilation with IDE/VScode.

  • Presets for qbitpp
  • Presets for chaiqpp

• VScode devcontainer mount the user's .ssh for syncing with gitlab.

  • no further setup required for syncing git, if the user already has set his ssh keys in $HOME/.ssh

⚠️ Breaking Changes

• operator_function API has changed to support both Hermitian and non-Hermitian operators; it now accepts an option map similar to that of iterative_eigensolver. See the New Features section for details.

🧪 Migration Notes

🚀 Highlights

• Patch release improving figure rendering and layout

🛠 Improvements

• Figure math rendering

  • LaTeX math is now rendered in axis labels and tooltips.

• Figure layout tuning

  • Improved default spacing between the figure title and the Y axis for better readability.

Development

🚀 Highlights

✨ New Features

• Introduced the Plotting module to create figures in Aleph for data analysis, including bar, line, scatter, and heatmap plots.
  • Supports legends, multiple series, multiple plots in the same figure, and a ton of customization options.
• Functions for the computation of matrix elements and expectation values with MPS and MPO.
  • matrix_element(bra,operator,ket) to compute the matrix element $\bra{\phi} O \ket{\psi}$ of an operator.
  • expval(operator,ket) to compute the expected value $\bra{\psi} O\ket{\psi}$ of the operator.
• Added accumulators for binning analysis of time series and markov chain data. Currently support accumulator("no_binning") and accumulator("binning"). The bin size can be set via accumulator("binning" ["bin_size" : 1024 ]).
• Added a FourBody symbolic fermionic operator: $c^{\dag}_{i}c^{\dag}_j c_k c_l$. An operator can be created via FourBody(i,j,k,l) where i,j,k,l are the sites where the operator acts.

🐛 Fixes

⚙️ Internal / Technical Changes

⚠️ Breaking Changes

🧪 Migration Notes

🐛 Known Issues

• Saving a figure file in a subfolder throws an error if the subfolder does not exists during the execution process.

🚀 Highlights

• Figures now support LaTeX math rendering across labels and annotations
• “Run” actions in Markdown now show only compatible agents

✨ New Features

• LaTeX math in figures
  • Figures can now render TeX math equations in titles, legends, tooltips, and other annotations.

🛠 Improvements

• Smarter agent filtering in Markdown
  • The Run action for Markdown code blocks now filters out agents that cannot perform direct executions.

🚀 Highlights

• Patch release improving error handling around execution flows

🛠 Improvements

• Job scheduling

  • Improved error handling and dialog closing behavior when scheduling jobs.

• Run now

  • Better error reporting and handling when running executions directly.

🐛 Fixes

• Reduced cases where errors could leave the UI in an inconsistent state.

🚀 Highlights

• Patch release fixing default timeout handling for Cloud Agents

🐛 Fixes

• Fixed handling of default timeouts for Cloud Agents to ensure executions behave as expected.

🚀 Highlights

• Job details are now easier to follow with unified and automatic refresh
• Clearer separation between personal and cloud agents
• Smoother code file creation flow from Markdown previews

✨ New Features

• Unified job refresh

  • Job details now have a single refresh action covering info, metrics, and outputs.
  • Job details automatically refresh every 30 seconds while the job is running.

• Improved job metadata

  • Job details now display the created at timestamp.

• Clear agent separation

  • Agent list now clearly distinguishes between personal agents and cloud agents.

• Smarter file creation from Markdown

  • When creating a file from a code block in a Markdown preview, the new file dialog now defaults to the Markdown file’s directory.

🐛 Fixes

• Minor UI and interaction fixes across job details and agent lists.

🚀 Highlights

• Cloud agents now report accurate system information
• Improved error reporting for cloud job scheduling failures

🛠 Improvements

• Accurate cloud agent system info

  • Cloud agents now correctly report CPU core count and total memory.

• Better cloud job error reporting

  • Jobs that fail due to cloud provisioning issues now surface clear failure states instead of hanging in scheduled or staled status.

🐛 Fixes

• Minor fixes to improve cloud execution reliability.

🚀 Highlights

✨ New Features

• Functions for the computation of matrix elements and expectation values with MPS and MPO.
  • matrix_element(bra,operator,ket) to compute the matrix element $\bra{\phi} O \ket{\psi}$ of an operator.
  • expval(operator,ket) to compute the expected value $\bra{\psi} O\ket{\psi}$ of the operator.
• Added accumulators for binning analysis of time series and markov chain data. Currently support accumulator("no_binning") and accumulator("binning"). The bin size can be set via accumulator("binning" ["bin_size" : 1024 ]).

🛠 Improvements

`

🐛 Fixes

⚙️ Internal / Technical Changes

⚠️ Breaking Changes

🧪 Migration Notes

🚀 Highlights

• Brand new Welcome screen with quick actions and direct access to docs and tutorials
• First preview of figure editing capabilities in Workshop
• Job metrics charts are now clearer and more accurate
• Small but handy file actions added

✨ New Features

• New Welcome screen

  • Quick actions to get started faster
  • Direct links to documentation and tutorials

• Figure editing (limited preview)

  • Edit selected figure properties after generation (title, legend, etc.)
  • Open the raw figure definition file for advanced edits

• File context menu actions

  • Copy path
  • Duplicate

🛠 Improvements

• Job metrics graph
  • CPU y-axis now adapts to observed samples instead of being capped at 100%
  • X-axis now stops at the last recorded sample to avoid empty tail space

🐛 Fixes

• Miscellaneous bug fixes and performance improvements across the Workshop UI.

🚀 Highlights

• CPU usage metrics updated to follow Linux conventions
• New Cloud Agents for Jobs offering dedicated, high-performance machines (limited preview)
• Minor bug fixes and performance improvements

✨ New Features

• Cloud agents for jobs
  • Jobs can now run on dedicated cloud agents with powerful machine configurations.
  • Designed for heavy, long-running, or performance-sensitive workloads.
  • Limited preview for now, we will open it up gradually

🛠 Improvements

• CPU percent metrics aligned with Linux standards
  • CPU usage now reports 100% = one CPU core, instead of a percentage of total machine CPUs.
  • Metrics are now consistent with standard Linux tools and expectations.

🐛 Fixes

• Minor bug fixes and general performance improvements.

⚠️ Breaking Changes

• CPU metrics interpretation has changed
  • CPU percentages now follow Linux semantics.
  • Jobs executed before this release will not be retroactively updated and may show inconsistent CPU values.

🚀 Highlights

• Point release focused on security and stability improvements

🛠 Improvements

• Security hardening

  • Tighter API security headers and CORS handling to better protect requests.

• Stability improvements

  • Improved backend reliability for smoother execution and job handling.

🐛 Fixes

• Minor fixes to improve overall robustness.

🚀 Highlights

• Clear separation between agent selection for direct executions and job scheduling
• Improved agent-related UI with clearer icons, tooltips, and guidance
• General UI polish

✨ New Features

• Separate agent selection flows

  • Direct executions and job scheduling now use distinct agent selectors.
  • Scheduling a job now requires explicitly selecting a job-capable agent.

• Agent-aware UI

  • New shared agent dropdown and icons used consistently across execution and job UIs.
  • Cloud agents now include an “About this (?)” menu entry to explain their behavior.

🛠 Improvements

• Smarter agent menus

  • Tooltips added to agents to clarify capabilities and constraints.
  • Cloud agents no longer show unsupported actions (reset, restart, troubleshoot, delete).

• Tooltip & UI polish

  • Tooltips and alerts now support multiline content.
  • Tree items now support tooltips for better discoverability.

⚠️ Breaking Changes

• Job scheduling now requires explicitly selecting a compatible agent.

⚙️ Internal / Technical Changes

• Minor update to the Agent Cloud Model to simplify alignement

🚀 Highlights

• Renamed Communal Agent to Interactive (Shared) for clarity
• Improved load balancing for shared executions
• Added in-product guidance for understanding cloud agents

✨ New Features

• Interactive (Shared) agent

  • The former Communal Agent is now called Interactive (Shared).
  • Provides a shared, cloud-hosted execution environment for interactive workloads.

• Cloud agent tooltip

  • Added a tooltip to explain the cloud agent model and when to use shared vs personal agents.

🛠 Improvements

• Better load balancing
  • Improved execution distribution on shared agents for more consistent performance.

⚙️ Internal / Technical Changes

• Refactored communal agents into a more general Cloud agent concept.

🚀 Highlights

• Communal agents are now available for direct executions (internal user only ATM).

✨ New Features

• Communal agents
  • A shared, load-balanced agent is now available for direct executions.
  • Communal agents are discoverable via the agents list for users who have access to it.
  • Designed for lightweight, shared execution without requiring an agent.

🎆🚀 Highlights — End-of-Year 2025 Release

• Markdown files editing now use word wrap by default for better readability
• Markdown preview gains support for figures, workspace images, and richer rendering
• Diff syntax is now highlighted and can be applied or reverted directly
• Multiple UI refinements and fixes across Kai

As we close out 2025, this release focuses on polish, usability, and smoother day-to-day workflows. Thanks for building with Kothar — and see you in 2026 🚀

✨ New Features

• Markdown word wrap (default)

  • Markdown files now open with word wrapping enabled for a more comfortable reading and editing experience.

• Enhanced Markdown preview

  • Render .figure outputs directly inside Markdown previews.
  • Render images stored in the workspace directly in Markdown.
  • Enables richer technical documentation and experiment notebooks.

• Diff syntax support

  • Diff blocks are now syntax-highlighted.
  • When sufficient context is available, diffs can be:
    • Applied directly
    • Reverted easily

🛠 Improvements

• Kai UI refinements
  • Multiple small UI improvements and bug fixes across the remembering, layout, and interaction flows.
  • More consistent behavior when working with diffs and previews.

🐛 Fixes

• Minor rendering and interaction bugs in Markdown preview.
• UI inconsistencies in Kai panels and interactions.

🎇 End of 2025 note
Thanks for being part of the Kothar journey this year. This final Workshop release of 2025 focuses on polish, stability, and better everyday ergonomics — setting the stage for an even more powerful 2026.

🚀 Highlights

• AI Assistant now produces shorter, clearer chat summaries
• Improved diff generation with stricter rules for cleaner, more unified patches
• AI Assistant can now create Markdown files in addition to Aleph code

✨ New Features

• Markdown file creation via AI Assistant
  The AI Assistant can now generate Markdown files directly, alongside Aleph source files.

🛠 Improvements

• Shorter AI chat summaries
  Chat summaries are now more concise and focused, making conversations easier to scan and review.

• Stricter unified diff generation

  • Tighter rules ensure diffs are more consistent and easier to apply.
  • Reduces malformed or overly fragmented patches.
  • Improves reliability when applying AI-generated changes.

🎄🚀 Highlights

• Execution metrics (CPU and memory) are now visible in both direct executions and jobs
• File tree renaming is now more reliable and easier to use (no more fighting the UI for focus)
• Kai can now answer questions about the currently open Aleph code

🎅 Happy holidays from the Kothar team!
Whether you’re writing code or running jobs — we hope this release makes your end-of-year workflows smoother.

✨ New Features

• Execution metrics visualization

  • Direct executions now display a summary of CPU and memory usage in the output.
  • Jobs now display CPU and memory usage over the full execution duration in the job details view.

• Automatic Aleph editor context in Kai

  • Kai now automatically understands the current Aleph editor context.
  • You can ask questions directly about the code you are working on — no copy/paste required

🐛 Fixes

• Reduced cases where rename inputs would lose focus or resize unexpectedly.

🚀 Highlights

• Workshop now supports displaying figures produced by Aleph modules
• The AI Assistant has been officially named Kai

✨ New Features

• Figure support in Workshop

  • Workshop can now display figures generated by compatible Aleph modules.
  • This enables direct visualization of computational results within the workspace.
  • Additional supported modules and formats will be announced later.

• AI Assistant naming

  • The AI Assistant has been baptized Kai.
  • This name is now used consistently across the Workshop UI and documentation.

🚀 Highlights

• AI Chats now support model selection (internal users only)
• First-class Markdown support added to the editor, including previews and executable Aleph code blocks
• AI Markdown rendering upgraded to match editor capabilities
• AI Chats now expose reasoning tokens

✨ New Features

• AI model selection (internal users only)
  Internal users can now choose which AI model to use when starting a chat.

• Markdown support in the editor

  • Create and edit Markdown files
  • Open live Markdown previews
  • Execute Aleph code blocks directly from Markdown documents

• Enhanced Markdown capabilities

  • LaTeX math rendering
  • Tables support
  • Syntax highlighting aligned with the main editor

• Reasoning tokens in AI Chats
  AI Chats now surface reasoning tokens when available, improving transparency during model interactions.

🛠 Improvements

• AI Markdown alignment
  • AI-generated Markdown now fully matches editor rendering capabilities
  • Consistent handling of math, tables, and syntax highlighting between AI output and the editor

🐛 Fixes

• Fixed UI layout issues that could cause AI Chats to spill over the main editor surface.
• Additional layout and spacing fixes across the Workshop UI for more consistent rendering.

v0.29.1

🚀 Highlights

✨ New Features

🛠 Improvements

🐛 Fixes

• Fixed lattice factories for oblique, rectangular and honeycomb geometries which were throwing errors for incorrectly defining the basis vectors.

⚙️ Internal / Technical Changes

⚠️ Breaking Changes

🧪 Migration Notes

🚀 Highlights

• Significant improvements to Aleph’s parallel-execution story: async now runs on a shared worker pool, and await_all, parallel_map, plus sleep round out the toolkit for structured concurrency.
• Model class has been revamped. Now as opposed to adding general functions via add_term(XX, neighbourhood) we have add(term([X, X], neighbourhood)). The add method is also more general, allowing for add(neighbourhood_rule(...)) and add(coefficient(...)).
• Model's add_observable method has been replaced with measure which takes a name and a term, (e.g.my_model.measure("name", Term(...)).
• Term and Coefficient have been introduced as helper types when defining model, with associated factories term and coefficient.
• Added new fermionic operators: NumberNumber which is a fused version of Number * Number, Number(List) a fused sum of number operators, NumberNumberNN a fused sum of NumberNumber operators for nearest-neighbours, and NumberNumberNNN a fused sum of NumberNumber operators for next nearest neighbours.

✨ New Features

• await_all(futures) waits for each Future in a list, consuming the inputs and returning the resolved values in submission order.
• parallel_map(values, mapper) applies the mapper concurrently across the shared async pool, keeping results aligned with the original list.
• sleep(seconds) pauses execution for a positive duration, throwing if a negative value is supplied.
• We added an eigensolver factory that picks the appropriate eigenvalue solver based on the properties of the input matrices, solves the eigenvalue problem and returns an object representing the solution. It currently supports standard and generalized eigenvalue problems for dense matrices (real, complex, Hermitian or not). The returned object has methods eigenvalues and eigenvectors to retrieve the eponymous quantities.
• We added an svd factory that computes the SVD decomposition of any dense matrix. It supports two algorithms: Jacobi and BDC. The default is Jacobi, which is more accurate for small and medium sized matrices, while BDC is faster for large matrices. The returned object has methods singular_values, matrix_u, and matrix_v to retrieve the eponymous quantities.
• We added a cholesky factory that computes the Cholesky decomposition of symmetric (Hermitian) matrices. It supports two algorithms: standard (LLT) and robust (LDLT). The default is standard, which requires the matrix to be positive definite. The robust algorithm (LDLT) can handle positive or negative semidefinite matrices. The returned object has methods matrix_l and matrix_u to retrieve the factors.
• We added an lu factory that computes the LU decomposition of any dense matrix. It supports two pivoting strategies: partial and full. The default is partial pivoting. The returned object has methods matrix_l, matrix_u, and matrix_p (and matrix_q for full pivoting) to retrieve the factors.
• We added a qr factory that computes the QR decomposition of any dense matrix. It supports three pivoting strategies: none, partial (column pivoting), and full. The default is no pivoting. The returned object has methods matrix_q, matrix_r, and matrix_p (if pivoting is used) to retrieve the factors.
• Add fast fourier transform (FFT) support with real-to-complex (r2c) and complex-to-complex (c2c) transforms
  • Support for 1D and 2D FFT operations via configurable axes parameter
  • Complex-to-complex transforms support forward and backward modes via fft and ifft functions
  • Real-to-complex transforms provide forward FFT only
• Added mpo(model) function to convert lattice models to finite MPO representations. This enables direct MPO construction from a Model definitions for use in DMRG and tensor network algorithms.
• Added observables_to_mpo(model) function to convert observables defined in a Model to MPO form, returning a map of observable names to their MPO representations.

🛠 Improvements

🐛 Fixes

• Container methods resize, insert, insert_at, replicate, replicated, extend and extended now clone inserted elements to match existing push_back, assignment, and constructor behavior.
• Restores the missing documentation for List::push_back in the API reference.
• Fixed a documentation bug regarding the factory functions make_random_mps and make_random_mpo, they were wrongly classified as functions and the first argument was missing from the documentations.

⚙️ Internal / Technical Changes

• Added copy constructors for BoundaryCondition types.

⚠️ Breaking Changes

• All container insertion methods now clone their inputs, so scripts that relied on resize, insert, insert_at, replicate, replicated, extend or extended mutating the original values will behave differently.

🧪 Migration Notes

🚀 Highlights

• System-level metrics are now collected during executions for better observability
• AI Chat assistant has been renamed
• AI Chat system prompt updated to significantly reduce hallucinations and improve API correctness

✨ New Features

• Execution system metrics collection
  • The agent now periodically collects:
    • CPU usage
    • Memory usage (Resident Set Size)
  • For direct executions, a summarized metrics report is sent at the end of the execution.
  • For job executions, all collected samples are streamed to the API periodically during runtime.

🛠 Improvements

• Improved AI Chat reliability
  • Updated system prompt to enforce stronger adherence to Aleph APIs.
  • Reduced hallucinations and off-spec responses when generating code or explanations.

🚀 Highlights

• Jobs are now fully manageable: create, edit, and delete
• UI improvements to job listing and duration display
• AI Chat updated to support the new backend API with streaming and cancellation (selected users only ATM)

✨ New Features

• Job management enhancements

  • Job names and notes can now be set and edited
  • Jobs can be deleted directly from the UI

• Job context menu
  A new context menu in the jobs list provides quick access to common job actions.

• AI Chat streaming & cancellation support
  The Workshop AI Chat now supports real-time streaming responses and graceful cancellation, matching the updated AI API.

🛠 Improvements

• Enhanced AI Chat behavior
  Improved interaction flow, responsiveness, and handling of intermediate states during AI generation.

• Improved job duration display
  Job durations are now normalized for longer-running jobs, making timelines easier to read and compare.

🐛 Fixes

• Fixed inconsistencies in how long-running job durations were displayed.
• Improved UI state handling during AI request cancellation.

🚀 Highlights

• Major refactor of the AI Chats API to improve reliability, extensibility, and alignment with the OpenAI API
• Streaming AI responses now supported for real-time content generation
• Backend now supports feature flags for safer rollouts
• New APIs added to update job names and notes

✨ New Features

• Streaming AI responses
  AI-generated content can now be streamed as it is produced, enabling more responsive user experiences.

• Client + server tool support
  The AI API now supports client-side tools in addition to server-side tools, unlocking more flexible AI workflows.

• Job metadata API
  New endpoints allow updating job names and notes

• Backend feature flags
  Feature flag support added in the platform (in addition to the Workshop) to progressively enable or disable features without redeploying.

🛠 Improvements

• AI API aligned with OpenAI’s canonical format
  The AI Chats API now uses a messages/roles structure consistent with the OpenAI API, simplifying integrations and future compatibility.

• Improved editor context handling
  Editor context is now explicitly passed and correctly processed by the model.

• Usage tracking per conversation

🐛 Fixes

• Streaming endpoints now return errors correctly instead of silently failing.
• Improved request lifecycle handling to prevent orphaned or hanging AI requests.

⚙️ Internal / Technical Changes

• Refactored AI tools container injection into a dedicated module.
• Added graceful cancellation of in-flight AI requests when the API shuts down (via AsyncDisposable callbacks).
• Client package updated to match the new AI API behavior and contracts.

⚠️ Breaking Changes

• AI Chats API has been significantly refactored
  • Request and response formats have changed
  • System prompt behavior is different
  • Streaming behavior is now first-class

⚠️ This will break the existing AI Chat feature in production until the Workshop is updated.

🚀 Highlights

• Support for uploading and viewing common file types, including PDF, PNG, GIF, JPG, TIFF, BMP, and WEBP

✨ New Features

• Preview for supported image formats, including zoom and pan
• Embedded PDF viewer with page navigation

🚀 Highlights

✨ New Features

• Added a new fermionic operator: NumberNumber which is a fused version of Number * Number.

🛠 Improvements

🐛 Fixes

• Prints Aleph's registered type names and the underlying exception message when an unhandled boxed exception escapes, restoring meaningful runtime error messages.

⚙️ Internal / Technical Changes

• Boundary conditions are now bound through a similar pattern to other enum types, appearing as static member of a global boundary_condition variable.

⚠️ Breaking Changes

• Boundary conditions are now denoted with the pattern boundary_condition.open, boundary_condition.periodic etc... as opposed to the old syntax, open_bc, periodic_bc etc...

🧪 Migration Notes

⚙ Internal / Technical Changes

• Improve file system interactions
• Improve tab dragging
• Add internal AI Chat POC

v0.27.1

🚀 Highlights

✨ New Features

🐛 Fixes

• Fixed a bug whereby two-element lists were implicitly converted to interval objects. This issue could lead to unexpected behavior.
  • Example: >>> [1,2] //-> Interval(lower=1.000000, upper=2.000000).

⚙️ Internal / Technical Changes

⚠️ Breaking Changes

• The Chebyshev series factories and fit function no longer accepts two-element lists as interval specifications. Users must now explicitly use the interval factory to define intervals.
  • Example: Instead of chebyshev_series_delta(10, [a, b]), use chebyshev_series_delta(10, interval(a, b)).

🧪 Migration Notes

🚀 Highlights

A clearer and more capable AI experience behind the scenes, with improved model flexibility and stronger Aleph code support.

⚙️ Internal / Technical Changes

• Internal users can now select which AI model to use.
• AI Chats now include Aleph code parsing and validation during code generation.

🚀 Highlights

• Added Model class which can be used to represent arbitrary lattice Hamiltonians and construct associated observables.
• The function chebyshev_series_delta and operator_chebyshev_series can be used together with iterative_eigensolver to compute interior eigenvalues and eigenvectors of operators. This is particularly useful for large-scale problems where traditional matrix representations are infeasible.
• Linear Algebra now has sparse matrices.
• Tensor multithreaded operations.

✨ New Features

• The matrix factory (which returns the matrix representation of an operator) now supports the as_sparsematrix type tag as an optional argument. This allows users to directly obtain a sparse matrix representation of an operator. Note that the type tag can be combined with other type tags such as as_real to specify the desired scalar type of the resulting sparse matrix (i.e. as_real | as_sparsematrix).
• The Model class can be used to represent a Hamiltonian and observables in terms of lattice coordinates.
• A new overload for operator_sum that takes an instance of Model and returns either a complex (default) or real OperatorSum depending on the type specifier (as_complex or as_real)
• The chebyshev_series_delta function creates a Chebyshev series that approximate the delta function over a specified interval. This is useful for spectral filtering and other applications requiring delta function approximations.
• The operator_chebyshev_series factory has been added to create operators based on Chebyshev series. This allows for efficient representation and manipulation of operators using Chebyshev polynomial expansions.
• Added more count methods to Fbit that help in counting the number of occupied sites on subsets and linear cuts of the system.
• Sparse matrices can be specified as a list of lists of numbers, similar to dense matrices but with the as_sparsematrix flag. One can also convert arrays and dense matrices directly to sparse matrices with the matrix factory. One can also create sparse matrices is by initializing them with the zeros, identity, uninitialized factory. Yet another way is to use insert which accepts three arguments: an array of row indices, an array of column indices, and an array of values. Lastly, one can of course use [] to extract elements or assign to them.
• Sparse matrices support addition, subtraction, multiplication (including with dense matrices), transposition, and conversion to dense matrices.
• Tensor multithreaded operations.
• Set to the detected hardware concurrency by default.
• The function set_tensor_threads(thread_number) can be invoked to control the thread count for tensor operations.
• Aleph global's control will override the tensor threadcount whenever it is called.
• Add extended for lists, an out-of-place extension of a list by an input list.
• Add as_real variants for TimerDuration conversion methods: hours, microseconds, milliseconds, minutes, seconds to retrieve more precise values when needed.

🛠 Improvements

• Add operator multiplication between OperatorSum and scalars with mixed coefficient types, enabling operations; real * ComplexOperatorSum and complex * RealOperatorSum.

🐛 Fixes

• stdout is now forced into line-buffered mode when Aleph runs without a TTY, ensuring CI/job executions emit work-in-progress logs immediately.
• time, benchmark_for and benchmark_for are now marked standalone, bringing the documentation to the Module Functions section.

⚙️ Internal / Technical Changes

⚠️ Breaking Changes

• The Polynomial class eval method has been renamed to evaluate for improved clarity and consistency.
• The Polynomial class evaluate method returns objects of the same type as the input argument. For example, if the input is a list, the output will also be a list.

🧪 Migration Notes

🚀 Highlights

• Introduction of API tokens to support secure automation and machine-to-machine integrations.
• Improved internal AI Chat tooling and a more secure SSE/EventSource implementation in the TypeScript client.

✨ New Features

• API Tokens — Enable automated workflows and machine-to-machine integrations. Each token is tied to a workspace and carries its own identity and permission scope.

⚙️ Internal / Technical Changes

• Internal users can now choose between multiple presets for AI Chat features.
• The TypeScript client package now uses a refactored SSE/EventSource implementation with improved security.

🚀 Highlights

• Improved onboarding flow for new users.
• Early groundwork completed for upcoming AI-assistant features.

✨ New Features

• Personal workspace initialization — New users now receive a pre-populated workspace on their first login, making it easier to get started immediately.

🐛 Fixes

• Resolved an intermittent issue where rate limits on job and user listings could cause partial results. A caching layer now reduces request pressure and prevents these failures.

⚙️ Internal / Technical Changes

• Pre-release of the AI Chats feature is now available internally for testing and validation.

🚀 Highlights

• Scalar-type-independent factory function names for improved ergonomics when building tensors.
• New tensor shape manipulation methods: transpose, flatten, and squeeze.
• Element-wise tensor-scalar operations: addition, subtraction, multiplication, and division.
• Unified API: random_tensor, filled_tensor, zero_tensor, delta_tensor, uninitialized_tensor.

✨ New Features

• Real and complex tensors now support .conjugate().
• Tensor Method transpose(tensor, permutation): permutes tensor dimensions with validation.
• Tensor Method flatten(tensor): reshapes tensor into a order 1 tensor.
• Tensor Method squeeze(tensor): removes all dimensions of size 1 from a tensor.
• Tensor-Scalar Operations: tensor + scalar, scalar + tensor, tensor - scalar, scalar - tensor for element-wise arithmetic.

🛠 Improvements

🐛 Fixes

• Improved numerical stability of DMRG with large numbers of sites.

⚙️ Internal / Technical Changes

• Re-implement tensor contraction to reduce compile times.

⚠️ Breaking Changes

• Tensor factory functions renamed: random_real_tensor → random_tensor, filled_complex_tensor → filled_tensor, etc...
  • All factory calls can specify the scalar type by parameters: as_real or as_complex (e.g., random_tensor([2, 3], as_complex)).
  • Default scalar type is complex; You must use as_real for real-valued tensors.

🧪 Migration Notes

⚙️ Internal / Technical Changes

• Add metrics for commands, actions and views

🚀 Highlights

• Random number generation and random distributions are now in Aleph.
• Aleph now supports creating random matrices using user specified distributions. Additionally, the default interval for randomly generated matrices is $[0,1]$ not $[-1,1]$.
• Native and binding-level exceptions now produce proper Aleph call-stack traces, making errors far easier to understand and debug.

✨ New Features

• random function can be used to generate random numbers on the fly.
• random_distribution factory can produce user specified distributions.
• random_generator allows users to seed their own random number generators.
• set_random and random_like can now take a Distribution specified by the user.
• Add functionality to extract the reduced density matrix of a many-body state.
• Add von Neumann and Rényi entropies functions.
• Automatic call-stack capture for any std::exception escaping AST evaluation.
• Consistent error formatting between native errors and script-level EvalError.
• Real and Complex tensor now support .conjugate() and conjugated() methods for complex conjugation. Leave the tensors unchanged in the real case.

🛠 Improvements

• Improved Library Reference: Operator types now appear consistently in the Types section with proper documentation, while constructors and internal operations are hidden from the Functions section.

🐛 Fixes

• Improved numerical stability of DMRG with large number of sites.

🚀 Highlights

This release focuses on UI polish and usability. Tree navigation is now more reliable in tight layouts, visual consistency is improved when renaming items, and several cross-platform rendering quirks—especially on Linux—have been addressed.

⸻

🛠 Improvements

• Tree View Responsiveness — Tree icons now stay visible even when the panel width shrinks, improving navigation in narrow layouts.
• Rename UI Consistency — Renaming an item now inherits the original item’s styling, making the interaction feel cleaner and more predictable.

⸻

🐛 Fixes

• Fixed vertical clipping of tree item titles.
• Fixed notification close button not appearing on Linux systems (likely due to font rendering differences).

🚀 Highlights

• Fix a bug that prevented some users to search in files
• Update agent reconnection logic

🛠 Improvements

• Agents now wait more predictably between connection instances, and give up after 3 days of unsuccessful attempts
• Agents logging is less verbose

🐛 Fixes

• Resolve a bug in search that prevented users to search in files if one of the file is empty

🧪 Migration Notes

• Agents will self-update automatically

⚙️ Internal / Technical Changes

• Add internal users-only feature

⚙️ Internal / Technical Changes

• Add internal endpoints for management

🧩 Fix

Commented include statements in Aleph scripts are no longer parsed as dependencies by the Agent. Example:

// include "dependency.aleph"

Previously, this line could incorrectly trigger dependency resolution and fail the execution. It is now correctly ignored.

🔧 Impact

This fix prevents incorrect dependency lookups and ensures consistent execution graphs when using commented or disabled includes.

We’ve added a simple, built-in way for anyone to report bugs — right from inside the app.

✨ New

• "Report a Bug" button in the main toolbar
• Submit issues directly without leaving the app
• Attach screenshots and annotate them — highlight problem areas or hide sensitive details
• Automatically includes helpful context (browser, version, environment)

🛠 Improvements

• Minor UI refinements and small stability fixes