# Changelog

All notable changes to this project will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## 3.0.26 - 29/07/25
- Added single tray galaxy reset option
- Bumped luwen from 0.7.5 -> 0.7.10
  - Chip detect now doesn't wait for eth to train for the 6U galaxy's, allowing multi tray resets to happen independently
- Updated readme with the new reset option

## 3.0.25 - 29/07/25
- Added packaging

## 3.0.24 - 04/07/25
- Now users have 2 galay reset modes available
  - glx_reset: resets the galaxy, informs users if there has been an eth failure
  - glx_reset_auto: resets the galaxy upto 3 times if eth failures are detected

## 3.0.23 - 03/07/25
- Bumped luwen 0.7.3 -> 0.7.5 to fix cargo lock compatibilty issue

## 3.0.22 - 02/07/25
- Bumped tt-tools-common 1.4.16 -> 1.4.17
- Bumped luwen 0.7.2 -> 0.7.3
- Bumped smi 3.0.21 -> 3.0.22

## 3.0.21 - 26/06/25

- Added option to not re-init chips after reset
- Updated galaxy 6u reset option from --ubb_reset to -glx_reset
- Removed the a3 arc message before doing a 6u reset, meaning we can reset even when chips are not pcie accessible
- Added eth link check and return failure if any of the eth links have a LINK_INACTIVE_FAIL_DUMMY_PACKET failure

## 3.0.20 - 04/06/25

- Chore - bumped tt-tools-common version to fix driver version check for compatability with tt-kmd 2.0.0

## 3.0.19 - 30/04/25

- Fixed an issue preventing the telemetry thread from being dispatched when the user clicked tab 2

## 3.0.18 - 22/05/25

- Added BH and WH UBB board type support
- Removed the dependency on tt-tools-common for this info

## 3.0.17 - 13/05/25

- Added proper telemetry heartbeat checks for Grayskull

## 3.0.16 - 12/05/25

- Used new ResetTypes from tools-common to simplify reset code
- Added a heartbeat spinner to the telemetry pane. We expect this spinner to update about twice per second. If the spinner is not moving, this indicates new telemetry is not being fetched.

## 3.0.15 - 24/04/25

- Patch for the ubb_reset to just discover local only post reset. Looks like eth port status 2 has been re-used to mean connected and pyluwen waits for it to clear, leading to eth timeout.

## 3.0.14 - 21/04/25

- Added wh ubb reset via command line `tt-smi --ubb_reset`. Intention is that this command line option will be removed and integrated into `tt-smi -r` after we update board detection with the correct external naming.
- Removed some unused imports and code - no functional changes

## 3.0.13 - 21/03/25

- Removed get\_sw\_versions

## 3.0.12 - 21/03/25

- Chore - bumped luwen version to include eth fw version check fix

## 3.0.11 - 13/03/25

- Chore - bumped luwen version to include enable chips with external connections but no routing

## 3.0.10 - 10/03/25

- Chore - bumped luwen version to include protoc lib detection check

## 3.0.9 - 07/03/25

- Chore - bumped luwen v

# Changelog

All notable changes to TT-Lang will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## Version 1.1.1

### Compiler

- Fix for live-interval boundary computation (issue [#536](../../issues/536))
- Fix for all-zero results in FP32 reductions (issue # [#533](../../issues/533))
- Fix for inferred `pop` and `push` (issues [#536](../../issues/536), [#554](../../issues/554))
- Fix for write pointer tracking on pipe sender accross iterations (issue [#578](../../issues/578))
- Fix to report data type mismatch error
- Fix to report DFB over allocation error (issue [#511](../../issues/511))
- Support for pipenet predicates `is_src`, `is_dst` and `is_active` (issue [#541](../../issues/541))
- Support for `ttl.math.typecast`

### Simulator

- Support for inferred `pop`, `push` and `copy`'s transfer handle `wait`
- Support for pipenet predicates `is_src`, `is_dst` and `is_active`
- Support `all_gather`
- Support `bfloat8_b`
- Improved/actionable error messages
- Improved performance by simulating math in FP32

### Infrastructure

- TT-Lang installable with `pip install tt-lang` for full installation and `pip install tt-lang-sim` for simulator only
- [Matmul benchmarks](benchmarks/matmul/README.md)

## Version 1.0.0

### Compiler

- Support `+=` syntax in conjunction with dot product (`@`) lowered to packer L1 accumulation
- Support implicit temporary compute-kernel-local DFBs
- Support `ttl.Pipenet`
- Support implicit `ttl.Block.push` and `ttl.Block.pop`
- Support implicit `ttl.Transfer.wait`
- Support for `expm1`, `exp2`, `ceil`, `sign`, `gelu`, `silu`, `hardsigmoid`, `square`, `softsign`, `signbit`, `frac`, `trunc` in `ttl.math`

### Simulator

- Support for `ttl.GroupTransfer`
- SPMD and mesh device simulation support
- Support for `ttnn.all_reduce` CCLs
- Use tracing to report statistics with `tt-lang-sim-stats`
- Remote L1 reads/writes statistics

### Examples and documentation
- Matmul tutorial

## Version 0.1.8

### Compiler

- Support for dot product operator (`@`) with lowering to [`ckernel::matmul_block`](https://docs.tenstorrent.com/tt-metal/v0.55.0/tt-metalium/tt_metal/apis/kernel_apis/compute/matmul_block.html)
- Support for fusing matmul and certain elementwise operations
- Support lowering to `pack_tile_block`
- Support for `ttl.math.fill`, `ttl.math.reduce_sum`, `ttl.math.reduce_max`, and `ttl.math.transpose`
- Support for arbitrary sub-blocking including dot product K-dimension to allow maximizing L1 usage and reuse
- Support for `sin`, `cos`, `tan`, `asin`, `acos`, `atan` in `ttl.math`
- Support for L1 sharded tensors
- Support for tensors with BF8 data type
- SPMD support (`ttnn.open_mesh_device`)

### Simulator

- Track L1 space and number of DFBs usage and warn when exceeded
- Support for tensors with row-major layout
- Support for L1 sharded tensors

### Examples and documentat

# Changelog

## [0.9.13] - 2026-10-02

### Changed

RTL simulator IP partitioning: one simulator run serving several chips via ip_layout.yaml.
RiscReset component interface.
Host memory channels set automatically based on topology.
IPMI reset improvements.

## [0.9.12] - 2026-09-28

### Changed

Simulator topology discovered through the silicon path.
IoOrdering parameter on Cluster read and write.
Quasar ATT address map.
Simulation servers whose host is gone are no longer reported.

## [0.9.11] - 2026-09-21

### Changed

TLBManager removed in favor of IoWindow; per-arch silicon TTDevice subclasses deleted.
Runtime telemetry and clock queries served by DeviceFirmware.
Optional cluster_id in the cluster descriptor.
Latest supported CMFW raised to 19.14.

## [0.9.10] - 2026-09-09

### Changed

Base API alignment: TTDeviceModel, DeviceFirmware, DmaInterface, IoWindow, MutexInterface, SystemMemoryAllocator.
SysmemBuffer ownership and host copy APIs; dma-buf export for RDMA.
Locks backed by KMD resource locks; tt-kmd-lib extracted.
Stateless NOC across TTDevice and SocDescriptor; xy_pair overloads removed.
Shared simulation server (sim_server) with client/host mode over sockets.

## [0.9.9] - 2026-07-09

### Changed

Blackhole runtime telemetry buffer accessors.
Precise ETH_LIVE_STATUS based APIs and Blackhole harvesting utility.
TTSim cluster descriptor discovered from simulator directory.

## [0.9.8] - 2026-07-01

### Changed

Per-op MMIO timeout for TLB-mapped device access.
set_power_state renamed to set_clock_state in TTDevice.
Simulation client mode over SimulationClient.

## [0.9.7] - 2026-06-26

### Changed

Device health errors returned from TopologyDiscovery.
Dedicated EthernetBroadcast and noc_multicast for remote chips.
DeviceTimeoutError exception type.
DRAM retrain on BIST failure.

## [0.9.6] - 2026-06-03

### Changed

ARM (aarch64) builds and fixes.
TTSimChip and RtlSimulationChip merged into SimulationChip, with multichip support.
TensixSoftResetOptions removed.
Warm reset with remote IO failure recovery.

## [0.9.5] - 2026-05-12

### Changed

Hardware hang detection for NOC and PCIe.
Tracy profiler integration with instrumentation across TLB, PCIe and sysmem paths.
DeviceProtocol ported to TTDevice, including DMA migration.
SocDescriptor split into static (SocArchDescriptor) and runtime parts.
LITERAL coordinate system in CoreCoord.
Multicast to all TENSIX cores.
SMN support.
SWEmuleChip software emulation chip and Quasar simulation support (incl. 4GB TLB).
Unified UmdException/UMD_ASSERT/UMD_THROW error handling across the codebase.

## [0.9.4] - 2026-03-18

### Changed

TopologyDiscoveryOptions refactoring.
TopologyDiscoveryOption to retrain ETH links on 6u.
TLBs for TTsim.
DRAM retrain support.
DeviceProtocol changes.
Simulator in TTDevice changes.
ETH heartbeat check.

## [0.9.3] - 2026-02-24

### Changed

Sigbus safe read write API.
Remove 4U related code.
Implement BH SPI as well, so full SPI support.
P150 expects harvested core

# Changelog

All notable changes to this project will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## 1.2.11 - 17/06/2025

### Updated

- Updated mesh coord generation to be connection type agnostic
- Added failure and exit if mesh type detected, but not enough connections
- Added warning in README about lack of supoort for BH and 6U boards

## 1.2.10 - 05/06/2025

### Updated

- Bumped tt-tools-common version to fix driver version check for compatability with tt-kmd 2.0.0

## 1.2.9 - 30/05/2025

### Updated

- Bug fix for https://github.com/tenstorrent/tt-topology/issues/39. Now the tool will use a DFS longest path to determine a linear layout if its not a fully connected graph.
- Updated initial device detection - now it needs full noc access for octopus and list options

## 1.2.8 - 08/05/2025

### Updated

- Fixed issue where tool would fail when PCI interfaces don't start from ID 0
- Now using actual PCI interface IDs from devices instead of assuming sequential numbering

## 1.2.7 - 07/05/2025

### Updated

- Use tools-common 1.4.15
- Use type checking in octopus reset

## 1.2.6 - 05/05/2025

### Updated

- Bug fix: added "ignore-eth" flag to first chip detect to avoid eth training loops forever and truly detect pcie only chips
- Chore: bumped luwen

## 1.2.5 - 15/04/2025

### Updated

- When flashing to isolated mode, we now flash the WH ethernet ports to a disabled state,
  in order to prevent their use.

## 1.2.4 - 02/04/2025

### Updated

- You can now run `tt-topology -l isolated` to flash cards to the default (non-connected) state
- Users are now warned about missing or loose cables

## 1.2.3 - 21/03/2025

### Fixed

- Bumped luwen (0.6.2 -> 0.6.3) to include eth version check bug for TG setup

## 1.2.2 - 13/03/2025

### Fixed

- Bumped luwen version to make it more robust against eth fw updates

## 1.2.1 - 13/03/2025

### Fixed

- Moved the spi reads after the reset to increase stability during M3 L2R copy
- Bumped luwen version

## 1.2.0 - 06/03/2025

### Fixed

- Updated how local eth board info is calculated to make it agnostic to eth fw version
- bumped tt-tools-common version
- Added traceback printing when catching exceptions in main.

## 1.1.5 - 14/05/2024

### Updated

- Bumped luwen (0.3.8) and tt_tools_common (1.4.3) lib versions
- Removed unused python libraries

## 1.1.4 - 25/03/2024

### Fixed
- Changed detect_chips with detect_chips_with_callback to enable detailed debug info.

## 1.1.3 - 22/03/2024

### Fixed
- Bumped tt-tools-common version to avoid pip discrepancy.

## 1.1.2 - 22/03/2024

### Fixed
- Fixed command line bug when no args are provided.

## 1.1.1 - 21/03/2024

### Fixed
- Fixed reference to pyluwen lib

## 1.1.0 - 12/03/2024

### Added
- Octopus Configuration (4 n150s connected to 1 galaxy)


## 1.0.2 - 12/03/2024

### Fixed
- Dependency bug with tt_tools

# Changelog

The **canonical, complete release log lives in [`debian/changelog`](debian/changelog)** —
that's the file the `.deb` packages are built from and where every release is
recorded in full. This file is a friendly pointer plus a summary of the most
recent releases; it deliberately does not duplicate the whole history.

To see everything:

```bash
less debian/changelog          # full history
git tag                        # released versions
```

## Recent releases

Releases 0.11.1 through 0.13.9 are recorded only in `debian/changelog`; this
list is a summary and is not kept for every release.

### 0.13.17
- **Fix**: review findings on the lanes and atmosphere. No false flares when stats drop
  out or lanes reorder, a fresh load clock after a reload, a forge that starts over
  per model, and trainer discovery moved off the render thread (with retries for
  a Python script that imports `ttml` late). The atmosphere's smoothed gauges
  restart when another model is featured, and a reused pid no longer keeps the
  previous training run attached or its cached `ttml` answer.

### 0.13.16
- **Change**: chips stay the ones a server holds open; `TT_METAL_VISIBLE_DEVICES`
  is not used to narrow them (it indexes the runtime's order, not the host's).
- **Add**: demo scenes `arcade-lanes` and `inference-music`.

### 0.13.14
- **Fix**: the MiniMax-Music3 audio server shows jobs in flight and finished
  requests (from `/v1/health`) and is labelled `MiniMax-Music3` (from
  `/v1/models`).

### 0.13.13
- **Fix**: a model server with no `/health` route (the MiniMax-Music3 audio
  server) shows as serving. The Inference view also asks `/v1/health`.

### 0.13.12
- **Add**: Arcade activity lanes. While a training run, a model load or a model
  serving is active, the top starfield rows show up to two lanes (name,
  progress, rate, `[t]`/`[i]` to open the full view); the stars tint by kind and
  flare per step or request.
- **Add**: an atmosphere behind the Inference view's snake: a forge while
  loading, a storm while a prompt is read, an aurora with comets, rain and a
  KV-cache tide while generating, a forming canvas for diffusion jobs.
- **Fix**: models launched from a weights path are labelled by their run
  (`rag-gate-27b`), not `model-dir` or `bundle`.

### 0.13.11
- **Fix**: on hosts that run the STIG or CIS audit rules, tt-toplike wrote
  4 to 6 GB of audit log per hour. The process scans opened `/proc/<pid>`
  files that only the owning user can read, and each refused open became an
  audit event. The scans now check ownership first and remember pids that
  refuse access. Thanks to Eugene Valkov, who reported it and wrote the first
  fix (issue 35).

### 0.13.10
- **Add**: Inference view (`i`) shows the port and chips each model runs on,
  for example `:8000 · chips 0,1`, on each roster row or in a header line for a
  single model.
- **Add**: `--tt-smi-reset-behavior` (`ignore`, `inform` default, `dazzle`,
  `demo`). A `tt-smi -r` in another terminal

# Changelog

All notable changes to this project will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## Unreleased

### Added

- `flash`: `--update-boot-images` writes the bundle's bootloader and recovery
  images even when the board already holds the same ones, for provisioning and
  board recovery.

### Changed

- `flash`: the boot-critical images (`cmfw`, `safeimg`, `safetail`, `failover`)
  and the ROM and failover descriptor tables are now left alone when the board
  already holds the same content, so a routine update no longer opens a
  power-loss window on the path by which a board boots at all. Whether an image
  is the same is decided by the SHA-256 and key hash that `imgtool` records in
  it, because signing is not reproducible: two builds of the same source differ
  only in the trailing signature. Pass `--update-boot-images` for the previous
  behaviour of writing them unconditionally.

### Fixed

- `flash`: a P300 chip running recovery firmware publishes no board id, so the
  pairing check filed it as not a P300, left its sibling alone in a group of
  one, and dropped both halves of the card from the flash list -- refusing the
  board because of the chip that most needed flashing. Such a chip is now
  identified by its PCI subsystem id, which carries the same UPI whatever the
  chip is running.
- `boot_fs`: `tt_boot_fs_fd.image_tag_str()` compared each `c_uint8` tag byte
  against the string `"\0"`, which never matched, so NUL padding was included in
  the decoded tag. Tags shorter than 8 bytes (e.g. `cmfw`) now compare correctly.
  This makes `read_tag` robust across the multi-table boot filesystem layout
  (ROM, failover, and mutable descriptor tables).

## 3.4.0 - 30/07/25

- Bump pyyaml 6.0.1 -> 6.0.2
- Improve error message formatting
- No longer have to use --force for flashing BH cards

## 3.3.5 - 03/07/25

- Bump luwen 0.7.3 -> 0.7.5

## 3.3.4 - 02/07/25

- Bump tt-tools-common 1.4.16 -> 1.4.17
- Bump luwen 0.6.4 -> 0.7.3

## 3.3.3 - 05/06/2025

- Bumped tt-tools-common version to fix driver version check for compatability with tt-kmd 2.0.0

## 3.3.2 - 14/05/2025

- Bump tt-tools-common version to latest

## 3.2.0 - 12/03/2025

### Updated

- luwen version bump to bring inline with tt-smi; provides stability fixes

## 3.1.3 - 06/03/2025

### Added

- luwen version bump to include bh arc init checks

## 3.1.2 - 28/02/2025

### Added

- Support for more BH cards: p100a, p150, and p150c

## 3.1.1 - 06/01/2025

### Updated

- Bumped luwen version to accomodate Maturin updates

## 3.1.0 - 29/10/2024

### Added

- Support for flashing the BH tt-boot-fs file format
- Bumped luwen version to 0.4.6 to allow resets when chip is inaccessible

## 3.0.2 - 17/10/2024

### Fixed
- Unbound variable when exception is thrown when getting current fw-version

## 3.0.1 - 16/10/2024

### Changed
- B

# Changelog

All notable changes to the TT-VSCode-Toolkit will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

---

## [0.1.29] - 2026-09-07

Correctness pass on what a TT-QuietBox 2 actually ships. The three-path venv
activation blocks added in 0.1.25 offered `~/.tenstorrent-venv` as the "QB2
pre-installed" environment for TT-NN and vLLM, and described a QB2 as "4
independent single-chip devices". Neither claim is true, and this release
corrects both everywhere they were found — including in this repo's own
`CLAUDE.md`, `llms.txt`, and two lessons outside the original diff
(`content/pages/FAQ.md`, `monkeypatch-ttnn.md`) that had never been touched
by the earlier topology/venv corrections this built on.

Verified against `tt-installer`'s `install.m4`, the published `tt-metalium`/
`tt-metalium-models` GHCR image configs, `tt-inference-server`'s `run.py`,
and tt-metal `main` — plus live end-to-end testing on a QB2-equivalent
board — rather than other Tenstorrent docs, which largely descend from the
same stale source note this release supersedes.

### Fixed

**`~/.tenstorrent-venv` is not a TT-NN or vLLM environment.** `tt-installer`
only ever installs `tt-smi`, `tt-flash` and (opt-in) `tt-topology` into it, so
`import ttnn` and `import vllm` both fail there. TT-NN lives **only** inside
the TT-Metalium container (`tt-metalium`, where `python3` is
`/opt/venv/bin/python3`); vLLM runs in a container `tt-inference-server`
launches. Corrected everywhere this was asserted or assumed:
- The "QB2 pre-installed image" activation line in all 11 TT-NN blocks across
  `explore-metalium`, `video-generation-ttmetal`, `animatediff-video-generation`,
  `cookbook-game-of-life`, `cookbook-mandelbrot`, `cookbook-particle-life`,
  `cookbook-image-filters`, `cookbook-audio-processor`, and the four lessons
  that used it as a premise (`lfs-00-intro`, `lfs-05-train-and-run`,
  `ct1-understanding-training`, `ct4-finetuning-basics` — their `ttml`-needs-a-
  source-tree conclusions were already correct; only the premise changed).
- `tt-installer`'s own "may not include the container wrapper" framing —
  backwards: a QB2 *has* the wrapper (it's the only way to reach TT-NN) and
  has no host-side TT-NN. Its "Test TT-Metalium" button and pytest demo
  example were also broken independent of this: missing `-c` (so `bash` tried
  to run the whole command string as a filename), a bare `ttnn.__version__`
  that doesn't exist on the built package (now `getattr(ttnn, "__version__",
  "import OK")`), and — for the demo — the wrong container and a Blackhole
  demo path that moved in tt-metal's January 2026 reorg (now
  `tt-metalium-models` + `models/demos/vision/segmentation/ufld_v2/blackhole`,
  with `--install-metalium-models-container` named as off-by-default).
- `vllm-production`'s new "On a QB2 — read this first" section (pointing a

# Changelog

All notable changes to this project will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## 1.4.17 - 02/07/2025

### Changed
- Loosened requirements on pyproject.toml to make it more compatible in different venvs

## 1.4.15 - 05/05/2025

### Changed
- parse\_reset\_json now returns a ResetInput with stricter typing

## 1.4.14 - 04/02/2025

### Added
- New flags in reset config file generation to disable sw\_version reporting

## 1.4.13 - 23/1/2025

- Removed nr\_hugepages count from compatibility, as hugepages allocation is tricky
  and deserves its own widget elsewhere.

## 1.4.12 - 16/1/2025

- Added TTHostCompatibilityMenu to replace Host Info and Compatibility boxes
- Added a count of nr\_hugepages to the TTHostCompatibilityMenu

## 1.4.11 - 30/12/2025

- Updated Luwen version to fix Maturin issue

## 1.4.10 - 16/12/2024

### Changed
- detect\_chips\_with\_callback now takes a print\_status arg

## 1.4.9 - 11/12/2024

### Changed
- A failed reset now results in a fail exit code on BH

## 1.4.8 - 11/10/2024

### Changed
- Updated reset completion logic to handle the case where the bmfw needs to upgrade itself

## 1.4.7 - 11/10/2024

### Added
- Implemented m3 reset option for Blackhole

### Fixed
- Fixed crash during driver version dection when the "extraversion" field is used
    - i.e. 1.28-bh

## 1.4.6 - 17/07/2024

### Added
- Reset support of Blackhole

## 1.4.5 - 11/07/2024

### Added
- Bump pyluwen library version (v0.3.8 -> v0.3.11)
- Moved pyluwen v0.3.11 to optional dependencies in pyproject.toml

## 1.4.4 - 21/06/2024

### Added
- Version bump of python dependencies in pyproject.toml (dependabot)
    - requests (2.31.0 -> 2.32.0)
    - tqdm (4.66.1 -> 4.66.3)
- Pydantic library version bump (1.* -> >=1.2) to resolve: [TT-SMI issue #27](https://github.com/tenstorrent/tt-smi/issues/27)

## 1.4.3 - 14/05/2024

### Added
- Arm platform check and warning for WH device resets in compatibility menu
- Added check for WH device init after reset and prompt user to reboot host if chips are still non recoverable
- Bumped textual (0.59.0) and luwen (0.3.8) lib versions

## 1.4.2 - 04/04/2024

### Added
- Added "silent" flag to WH and GS resets to make them more versatile for use in other tools

## 1.4.1 - 22/03/2024

### Fixed
- removed pyluwen version to avoid dependency issues in other repos

## 1.4.0 - 19/03/2024

### Added
- detect_device_fallible that will provide feedback about chip state during init

### Fixed
- Update min driver version to 1.26 to perform lds reset
- Reset config file uses dev/tenstorrent id
- Catch JSON errors in reset config parsing
- Make nested dirs when initializing reset config path

## 1.3.0 - 06/03/2024

### Added
- Migrated GS Tensix reset to tools_common
- Migrated all related GS data files
- Functions to fetch arc and eth fw versions from telemetry

# Changelog

All notable changes to this project will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## 0.2.2 - 31/07/2024

### Added
- Added glx reset support
- Threaded start and end of burnin to increase burnin speed
- Added prints to indicate which chip we are currently running on
- Added support for bh harvesting

## 0.2.1 - 16/01/2024

### Bug fix
- Fix for https://github.com/tenstorrent/tt-burnin/issues/6
- BH reports asic temperature as a signed 16_16 int unlike GS and WH
- Added missing support to report BH asic temperatre

## 0.2.0 - 29/10/2024

### Added
- BH burnin support

## 0.1.1 - 14/05/2024

### Updated

- Bumped luwen (0.3.8) and tt_tools_common (1.4.3) lib versions

## 0.1.0 - 04/04/2024

First release of opensource tt-burnin

### Added
- GS and WH burnin support

# Changelog

All notable changes to this project will be documented here.

## 2.0.0.dev0

- `get_device_name` gives a Quasar device the Blackhole names (`P150`, `P300`,
  `P150x4`, `P150x8`, `BHGLX`), so code keyed on those names runs on Quasar.
- Establish `tt-transformers` as the standalone package for TTTv2 modules,
  runtime components, sampling, and twelve experimental model families.
- Add reproducible package builds and CPython 3.10/3.12 host validation.
- Add serialized Wormhole and Blackhole hardware qualification policy.
- Require `ttnn==0.79.0`. Trace capture now acknowledges its allocations to
  ttnn's trace allocation tracker (tenstorrent/tt-metal#53735).
- Rename the warmup override `trace_prefill_warmup_seq_lens` to
  `prefill_warmup_seq_lens`.
- Remove the TTTv1 sampler surface from `tt_transformers.sampling`
  (`generator`, `tt_sampling`, `tt_penalties`). Use
  `tt_transformers.modules.sampling`.
- With any trace mode, a prefill bucket that has no captured trace is served
  eager instead of failing. The first eager request per bucket logs one
  warning, and the load logs which buckets are traced.
- A traced configuration whose warmup lengths leave a servable prefill bucket
  uncompiled is refused at load with a `ValueError`.
- A `max_seq_len` that isn't a prefill bucket boundary now warms the bucket it
  pads to.
- Warmup compiles the prefix-cached variant of every prefill bucket a cached
  prompt can reach, including the top one. A configuration whose `max_seq_len`
  is itself the top bucket gains one such case per sampling path, captured as a
  trace where that bucket is traced; the Llama-3.1-8B N300 trace region grows
  to 70 MB for it.
  The Qwen3-32B T3K trace region grows from 120 MB to 150 MB: its batch-32
  cases gain three prefix-cached 1024 traces and need up to 125 MB.
- Default KV-cache sizing leaves room for the decode page table. A paged KV
  cache narrower than the decode page table is refused at load with a
  `ValueError`, instead of failing on the first decode step.
- On-device greedy sampling breaks an exact logit tie by the lowest token id,
  as host argmax does. Before, the pick depended on the batch slot and could
  vary between runs.
- vLLM serving of Qwen2.5-72B (T3K) and Qwen2.5-7B (N300) prefills prompts in
  the 128-token bucket in fixed waves of four rows, padding a partial wave, and
  warms exactly that shape. Under many concurrent short prompts, time to first
  token falls by more than half against one prompt per pass; a lone short
  prompt waits 85–106 ms longer. Greedy output on these prompts is
  deterministic but is the four-row pass's text, so it differs from one-row
  prefill and from TTTv1: pinned greedy texts for these two models move once.

This is a developer preview. No model is promoted beyond the support status in
`SUPPORT.md` and its example manifest.

# Changelog

All notable changes to TT-Bio are recorded here. Versioning is [SemVer](https://semver.org);
releases are cut from a commit that has passed the on-hardware test suite (see `RELEASING.md`).

## [Unreleased]

## [0.26.1] - 2026-10-29

### Fixed
- **ESMFold2 no longer runs out of device memory on a worker that has folded many different inputs.** A worker keeps
  its chip open across jobs, and the compiled programs for every input shape it has folded stay in device memory.
  After about ten hours of mixed lengths a Wormhole chip had too little room left for a 722-residue fold, and JapanFold
  failed up to 65 % of such tasks with "Ran out of device memory". A fold that is refused memory while programs are
  cached now drops them and runs again; folds that fit are unchanged. Seen on JapanFold with 0.25.0 and 0.26.0.

## [0.26.0] - 2026-10-29

### Changed
- **ESMFold2 and ESMFold2-Fast normal mode on Wormhole run the token attention as one fused program.** A 512-residue
  fold on a Wormhole Galaxy chip drops about 0.85 s (ESMFold2 25.9 -> 25.0 s, ESMFold2-Fast 16.0 -> 15.1 s). Structures
  change at the precision level: graded over 44 paired folds per model in normal mode, accuracy and docking success
  unchanged. Blackhole and `--fast` already ran it.
- **BoltzGen normal mode on Wormhole conditions its diffusion steps ahead of time.** A 514-token design on a
  Galaxy chip at 1000 MHz drops from 43.3 to 40.9 s. Designs change at the precision level: graded over 64 paired
  designs, design pTM, ipTM and complex pLDDT unchanged within the seed-to-seed spread.

## [0.25.0] - 2026-10-28

### Changed
- **BoltzGen's token diffusion transformer does less attention work.** In normal mode its pair bias was padded to
  768 tokens for a 544-token design; it now pads to 576, and on Blackhole the attention runs as one fused program
  in both modes. A 514-token design drops about 1.2 s on a Wormhole Galaxy chip (46 -> 45 s) and on a Blackhole
  chip about 2 s in normal mode (22.9 -> 21 s) and 1.5 s with `--fast`. Designs change at the precision level:
  graded over 64 paired designs per arch in normal mode and 32 with `--fast`, confidence metrics and filter pass
  counts unchanged. Other models keep their padding.
- **BoltzGen normal mode on Blackhole replays the diffusion step from a device trace by default.** Output is
  identical, the 1831-residue size ceiling holds, and with the triangle-multiplication change above a 514-token
  design on a p300 chip drops from 25.2 to 23.0 s. `--no_diffusion_trace` restores the eager step. `--fast` replays it too, and
  computes each step's conditioning inside the trace instead of several steps ahead: a 514-token fast design on a
  p300 chip drops from 32.4 to 23.3 s on a host-bound machine.

## [0.24.0] - 2026-10-27

### Changed
- **RoseTTAFold3 on Wormhole runs the fast triangle and transition kernels in normal mode.** A 512-residue fold
  on a Galaxy chip drops from 55.2 to 38.1 s. Structures change at the precis

# Changelog

All notable changes to TT-Atom are recorded here. Versioning is [SemVer](https://semver.org);
releases are cut only from a commit that has passed the on-hardware release gate — accuracy
parity, no OOM across the supported size range, no perf or UX regression, and a clean install
smoke (see `RELEASING.md`).

## [Unreleased]

Behaviour changes are all in the knobs and the caches, not in the models: every numeric path is
byte-identical to 0.3.0.

### Fixed
- The release gate's perf leg no longer decides by luck. It took one measurement per model against
  a fixed 15% threshold, and on a p150a the throughput spread between runs reaches that threshold
  on its own: in the gate run that closed this entry, `uma-s-1-omol-batch` drew 88.9, 104.4 and
  104.8 sys/s, and the first of those alone is a 20% shortfall against the baseline. The leg now
  waits for every other process to let go of a card and gates the median of three independent
  runs, printing all three. With no quiet window it measures anyway and reports a shortfall as
  `GAP`, since a contended run is not evidence of a regression.
- `benchmarks/_harness.host_quiet()` reported the host busy forever. It grepped process command
  lines for `tt_bio`, which matches any agent whose own arguments merely mention it; it asks the
  kernel who holds a `/dev/tenstorrent` node now. The three benchmarks that wait for a quiet host
  used to burn their full 40-minute budget and stop without measuring.
- The edge-bucketing speedup in the README and `docs/orb-port.md` is re-measured and now names the
  environment it was taken in. It said 1.4x cold wall-clock on a 20-system screening stream; two
  draws on the pinned tt-metal source build give **1.11x and 1.14x** (235.1 / 236.5 s unbucketed
  against 210.9 / 207.6 s bucketed). What bucketing saves is compiles, and that is unchanged and
  exact: 20 distinct edge shapes collapse to 7 buckets and 2350 fewer kernel files are built,
  identical to the file across both draws. The wall-clock ratio fell because a compile costs about
  half what it did in whatever environment produced the earlier number — which that log does not
  record, so the three subprocess benchmarks now stamp `ttnn_version` next to `git_sha`.
- `benchmarks/_harness.sandbox_env` resolves the sandbox `$HOME`. A relative `--workdir` reached
  the child as a relative `$HOME`, and tt-metal resolves that against the child's own working
  directory: it died deep in the JIT build with "Failed to open compile failure log file" and a
  path that reads as correct. `bench_compile_pain.py` also defaulted to `--card 3`, a card that
  does not exist on the host this repo is routed to.
- `TT_ATOM_SCATTER_THRESHOLD` is documented. It is the node count above which UMA's dense one-hot
  scatter gives way to the linear path, and therefore what bounds DRAM on a large cell, but it
  appeared in no doc.
- The release gate reads `OVERALL: PASS` again. Both Orb perf rows were seeded on a stock `ttnn`
  0.68.

# Changelog

All notable changes to tt-forge-compiletron are documented here.

## [Unreleased]

### Added
- `docs/kv-cache-bench.md` — teaching companion for the StaticCache KV cache
  benchmark, explaining the two-graph pattern and why static shapes matter

---

## [1.6.0] — 2026-06-30

### Added
- **StaticCache KV cache decode benchmarking** — `bench_decode.py` now compiles
  a second forge graph for the decode step using `transformers.StaticCache`.
  The StaticCache is embedded in `KVDecodeWrapper` as a submodule so forge
  traces K/V tensors as model state and emits `FillCache`/`UpdateCache` ops.
  Falls back to full-recompute for models that don't support `cache_position`.
- `_try_kv_decode()` function — detects model dtype to avoid bfloat16/float32
  mismatches, resolves tokenizer from loader or AutoTokenizer, pre-fills cache
  on CPU before forge compilation.
- Bestiary `decode_note` field now records the method used per model
  ("StaticCache KV cache" vs "no KV cache — full recompute per step").

### Changed
- Decode results updated for all 5 stages — GPT-2 2.30→5.52 tok/s, OPT
  3.98→5.05 tok/s, Phi-2 1.48 tok/s (new), Falcon 3.30 tok/s (new),
  LLaMA-LoRA 2.86 tok/s (new), Gemma-LoRA 2.40 tok/s (new), and more.

---

## [1.5.0] — 2026-06-30

### Added
- **`scripts/bench_decode.py`** — dedicated LLM decode benchmark measuring
  TTFT, prefill tok/s, and decode tok/s for all compiled causal LMs.
  Subprocess isolation + tt-smi health check prevent hardware lockups.
- **Leaderboard columns** — TTFT, Prefill tok/s, Decode tok/s, Params (M)
  replace the old Infer p50 / Throughput columns in `docs/leaderboard.html`.
- 5 benchmark stages: Stage 1 (GPT-2, OPT), Stage 2 (Phi-2, BLOOM, CodeGen),
  Stage 3 (Falcon, Allam, LLaMA-LoRA, Gemma-LoRA), Stage 4 (Qwen 2.5,
  Phi-1 LoRA), Stage 5 (DeepCogito, DeepSeek Coder, frontier models).
- `params_m` field added to all benchmarked bestiary entries.
- `hf:` loader prefix for frontier HuggingFace models loaded without a
  tt-forge-models seed loader.

### Changed
- Bestiary `throughput_unit` relabeled from generic `tok/s` → `prefill_tok/s`
  for all 54 causal LM entries to prevent confusion with decode throughput.

---

## [1.4.0] — 2026-06-29

### Added
- **`scripts/install.sh`** — turn-key smart installer: hardware pre-check,
  hugepages, disk space, forge venv, XLA venv, mesh descriptor probe,
  tt-forge-models clone, stale-shm cleanup. Outputs color-coded summary table.
- **RAM/DRAM budget calculator** — skips models whose weights exceed available
  system RAM + per-chip DRAM; prevents OOM crashes at load time.
- **`scripts/setup-venvs.sh`** — minimal venv setup script for clean Ubuntu
  24.04 installs on Tenstorrent Blackhole hardware.
- Self-contained patches directory — tt-forge-models fixes applied at
  expedition startup without modifying upstream.
- `--ephemeral` / `--evict-failures` flags — evict HF weight cache after
  each model to reclaim disk space on small-storage machines.

### Changed

# Changelog

All notable changes to tensix-viz are documented here.

## [1.3.2] - 2026-09-28

### Fixed

- **Cluster views no longer collapse to their caption's width**
  (`src/cluster.js`). Cluster tiles are empty divs with no width of their own,
  so a shrink-to-fit parent sized the whole widget to its one-line spec text —
  Galaxy SC's 128 chips rendered as 2.7px dots. The tile grid now gets a
  preferred width from its column count (32px tiles, or 10px dots above 64
  chips) and `max-width: 100%`, so it still shrinks on phones.
- **Zoomed cluster breadcrumb no longer covers the spec caption**
  (`tensix-viz.css`). `.tv-cluster.tv-zoomed-in` reserves room at the top for
  the "Cluster › Server N" breadcrumb.

### Docs site (landing page)

- Hero and two-column sections now stack at 1120px instead of 900px. Between
  those widths the QB2 hero chips were squeezed to ~120px and the Memory code
  block hid up to ~140px of each line; the hero now gets full 340px chips there.
- Code blocks wrap at ≤800px instead of hiding most of each line in a sideways
  scroll box (the Quick-start block hid ~400px per line on a phone).
- Phones: tighter padding around the hero system (chips 85px → 119px at
  375px wide), and `kernel_dispatch` may wrap so the mode table fits.

## [1.3.1] - 2026-09-28

### Fixed

- **Chips in a card/system no longer come out at different sizes**
  (`src/card.js`, `src/chip.js`). `TensixViz` caps its drawing size to its
  parent's `clientWidth` once, at construction. `CardViz` builds chips one at a
  time into a flex row, so each chip measured the room its predecessors had
  left — on the docs hero, QB2's second chip in each card rendered at roughly
  half the size of the first. Card chips now pass the new
  `fitContainer: false` option and keep their full logical size; CSS fits them
  (`.tv-chip-wrapper { flex: 0 1 auto; min-width: 0 }`).
- **Shrunk canvases are no longer squished.** The inline `style.height = Hpx`
  overrode `.tv-chip-wrapper canvas { height: auto }`, so whenever
  `max-width: 100%` narrowed a canvas its height stayed put (aspect ratio 0.8
  instead of 1.42). Canvases now get `max-width: 100%`, `height: auto` and an
  explicit `aspect-ratio`, so they scale cleanly after construction too.
- `.tv-card`, `.tv-system`, `.tv-card-wrapper` and `.tv-cluster` can shrink to
  their container (`max-width: 100%` / `min-width: 0`).

### Docs site

- Hero and "chip anatomy" grids use `minmax(0, 1fr)` tracks; a bare `1fr`
  couldn't shrink below the Memory section's `<pre>`, which pushed the page
  sideways even on desktop.
- The mode reference table scrolls inside its own box on phones.
- Hero theme buttons are grouped with the canvas they theme (they were a stray
  third grid cell under the hero text).
- Examples: demo cards can shrink below their canvas's preferred width, page
  padding scales down on phones, and the theme toolbar wraps.

## [1.3.0] - 2026-09-24

### Added

- **`setActivity(value)` and `setProgress(value)`**

Booth — Changelog
=================

## Booth 0.6.1

### Backends

- `--nvidia-cubin` puts every kernel of a module into the one cubin, so
  ggml-cuda files with several kernels no longer stop at E600
  (Zane Hambly, 2026-10-26)

- `--nvidia-cubin` lowers local loads and stores, half and bfloat16
  conversions, rounding, votes, membar, nanosleep, 16-bit parameters and
  constant divisors, and sizes each kernel's registers from its own counts
  (Zane Hambly, 2026-10-26)

- `--nvidia-cubin` holds a register only where its value is live, places join
  barriers by the blocks a region covers, and lowers wmma load, store and mma
  to scalar SASS, so topk-moe and lightning-indexer build (Zane Hambly,
  2026-10-26)

- `--nvidia-cubin` lowers cvta.local and cvta.global, so cumsum and quantize
  get as far as their vprintf call (Zane Hambly, 2026-10-26)

- `--nvidia-cubin` spills values to the local frame when a kernel needs more
  registers than SASS has, so mmvq gets past E603 (Zane Hambly, 2026-10-26)

- `--cuda-host` and `--cuda-exe` emit the host half of a CUDA compile, so kath
  builds a whole program the way nvcc does: the source back out with the device
  definitions removed, each launch through `__cudaPushCallConfiguration` and a
  stub calling `cudaLaunchKernel`, the kernels registered from a constructor out
  of a fatbin carrying both the cubin and the PTX, and the system C++ compiler
  driven to link it. Clean-room `cuda.h`, `cuda_runtime.h`, `cuda_fp16.h`,
  `cuda_bf16.h` and `cublas_v2.h` under `runtime/include` (Zane Hambly,
  2026-10-26)

- `--cuda-host` writes each kernel's stub where the kernel stood and keys a
  launch on the instantiation rather than its position, so a kernel launched
  from inside a host template or through a macro reaches the right one, and it
  spells enum, struct, vector and function template arguments, so ggml-cuda
  files get through the host pass (Zane Hambly, 2026-10-26)

- `--cuda-host` writes out a variadic template argument pack, spells a function
  template-id argument rather than the symbol it lowered to, keeps a launch
  whose callee ends in `>>`, registers up to 4096 kernels and refuses with E622
  above that instead of falling silent, and `kath cuda` finds Booth's own
  `runtime/include` from the binary rather than the working directory, so
  binbcast and cpy get through (Zane Hambly, 2026-10-26)
- `--cuda-host` carries const on the BIR type itself, so a kernel taking
  `const float *` registers the instantiation its launch actually names
  rather than one spelled without the qualifier (Zane Hambly, 2026-10-26)

- `kath cuda -c` compiles one `.cu` to one relocatable object rather than a
  whole program, with `-fPIC` for a shared library, each object carrying its
  own fatbin and registration constructor the way nvcc does, so many
  translation units link into one library (Zane Hambly, 2026-10-26)

- `kath cuda` takes `-Idir` written up against the flag as well as `-I dir`,
  rather than dropping it and c

# Changelog

All notable changes to the Tenstorrent Simulator Playground will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## [1.4.0] - 2026-01-27

### Added

- **Updated Demo Assets**
  - New digit recognition demo GIF with improved visualization

### Fixed

- **Docker Desktop Compatibility**
  - Fixed 500 Internal Server Error when running digit recognition in Docker containers
  - Implemented native Docker inference that runs Python directly instead of attempting WSL calls
  - Proper environment variable handling for containerized ttsim execution
  - Automatic SOC descriptor selection for Wormhole/Blackhole chips in Docker

## [1.3.0] - 2026-01-27

### Added

- **Enhanced Network Visualization**
  - 28×28 pixel grid visualization for input layer displaying the actual drawn digit
  - Vertical output column showing all 10 digit classes (0-9) with labels
  - Green ring highlight indicator for the predicted digit
  - Pixel data pass-through from drawing canvas to network visualization

### Changed

- **Performance Statistics Display**
  - Average throughput now displays with 2 decimal places for precision

## [1.2.0] - 2026-01-27

### Added

- **Digit Recognition Page**
  - Interactive drawing canvas for handwritten digit input (0-9)
  - Real-time MNIST neural network inference on Tenstorrent simulator
  - Network architecture visualization showing layer activations and connections
  - Performance statistics tracking (average latency and throughput)
  - Confidence display with horizontal bar chart for all 10 digit classes
  - Trained 2-layer MLP model (784→128→10) achieving ~98% accuracy
  - New digit recognition API endpoints (`/api/digit/predict`, `/api/digit/model-info`)

- **Simple 2-Layer MLP Operation**
  - Configurable neural network architecture editor
  - Interactive network diagram showing input, hidden, and output layers
  - Configurable sizes: input (32/64), hidden (32/64), output (16/32/64), batch (16/32/64)
  - Parameter count display with real-time updates
  - Visual representation of layer connections and ReLU activation

- **Matrix Multiplication Operation**
  - Support for 32×32 and 64×64 matrix operations
  - Optimized for TTNN performance benchmarking
  - Added matrix multiply icon to operation selector

- **React Router Integration**
  - Separate pages for Digit Recognition (/) and Mathematical Operations (/math-operations)
  - Clean URL structure with browser navigation support
  - Navigation sidebar with page links

### Changed

- **UI Redesign**
  - Non-collapsible sidebar with fixed width (256px) for better text display
  - Simplified branding: "TTSim Playground" with chip icon
  - Mathematical Operations page header now matches Digit Recognition style
  - Operation selector reorganized into 4×3 grid by category
  - Updated icons for subtract (circle with line) and matrix

# Changelog

Notable changes per release. Format loosely follows
[Keep a Changelog](https://keepachangelog.com/en/1.1.0/);
this project does not yet promise SemVer compatibility on the RPC
wire format or library API surface (we're not 1.0).

## Unreleased

V2 virtio-dispatch redesign. The kick ring + completion ring + host-
side throttle that grew up around #184 are gone; in their place is a
per-(slot, queue) dirty bitmap in BRISC L1. The bitmap is level-
sensitive — guest QUEUE_NOTIFY storms coalesce into a single set
byte, so the dispatch path can't fall behind under any burst. Wire
incompatible with 0.9.0; `TENSIX_PROTOCOL_VERSION` bumped 4 → 5.

### Added

- **V2 dirty-bitmap dispatch** (`#187` / `#188` / `#189`). BRISC
  writes 1 to `CTRL_OFF_DIRTY[slot][queue]` on every guest
  QUEUE_NOTIFY; the daemon's `Dispatcher` clears the byte and
  dispatches each pass. Replaces V1's 2048-entry kick ring +
  daemon-side `consume_kick_ring_pass` consumer.
- **V2 processed-cursor table** at `CTRL_OFF_PROCESSED`. Daemon
  publishes `used.idx` after each successful dispatch so
  warm-resume reads cursors directly without re-probing guest
  DRAM.
- **`bhx_notify_events_total`, `bhx_dispatch_passes_total`,
  `bhx_dispatch_queues_drained`** Prometheus counters surface the
  new dispatch path. The burst regression test (`scripts/
  soak_virtio_burst.py`) asserts `dispatch_passes_total > 0` to
  confirm the workload reached the new path.
- **`scripts/soak_virtio_burst.py`** — multi-queue burst regression
  test. Sustains 16-job direct=1 fio randwrite + a tight
  `printf` loop to `/dev/console`, samples `/metrics` every 1 s,
  and verifies the daemon log contains zero
  `kick.*drop|rescue|throttle.*ENGAGE` matches.
- **`DaemonState.chip_reset_this_session`** flag — gates
  `maybe_opportunistic_reset_board` so 4-way parallel cold boots
  reset the chip exactly once, not once per L2CPU. Without this
  the second-and-later resets blip the chip while earlier-booted
  L2CPUs hold mmap pages, SIGBUSing their workers.
- **`Dispatcher` (was `KickPoller`)** with documented testability
  seam (`CtrlL1Access` trait); `drain_dirty_bitmap` is unit-tested
  against an in-memory L1 fake covering all five visit/clear
  semantics cases plus the address-formula pins.

### Changed

- **`KickPoller` → `Dispatcher`**, plus `kick_poller` → `dispatcher`
  field on `DaemonState`, `tensix-kick-poller` → `tensix-dispatcher`
  thread name, `[kick-poller]` → `[dispatcher]` log tag,
  `kicks_consumed` → `dispatches_total`,
  `last_kick_slot_queue` → `last_dispatch_slot_queue`. Pure
  rename; no behavior change. V1 vocabulary scrubbed throughout
  the codebase (firmware, daemon, scripts, docs).
- **`CTRL_SIZE` shrinks 36 KiB → 4 KiB**. V2 footprint is ~1.5 KiB;
  the rest is reserved for future fields.
- **Stats-page offsets repacked** — V1 `STATS_OFF_KICK_DROPS`,
  `STATS_OFF_COMPL_EVENTS`, `STATS_OFF_LAST_COMPL` retired with
  V1 (#190); deprecated PRECAP / BLINDCAP / POSTCAP slots dropp

# Changelog

All notable changes to this project will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## [0.1.6] - 2025-01-14

### Added
- **Memory Configuration Support**: New command-line options for tensor memory configuration
- `--memory-config CONFIG`: General option with choices `[dram, l1]`
- `--dram`: Shortcut flag for DRAM memory (default)
- `--l1`: Shortcut flag for L1 memory
- Memory configuration extraction from CSV profiler output
- Memory config display in test result summaries

### Changed
- **Default tensor shape reduced from `[1, 1, 1024, 1024]` to `[1, 1, 32, 32]` for better performance**
- Enhanced `create_test_tensor()` function to accept memory_config parameter
- Updated all `ttnn.from_torch()` calls to use memory_config parameter
- Improved CSV extraction to read memory configuration from profiler output
- Enhanced debug output to show memory configuration

### Technical
- Added `validate_memory_config()` function with alias support
- Extended environment variable system with `TTPERF_CUSTOM_MEMORY_CONFIG`
- Updated operation_configs.json to include memory_config field
- Enhanced test file configuration parsing to handle memory settings
- Improved result reporting to include memory configuration details

## [0.1.4] - 2025-01-14

### Changed
- **Major Improvement**: Configuration extraction now reads from CSV profiler output instead of parsing text with regex
- Replaced 50+ complex regex patterns with structured CSV data parsing
- Enhanced `extract_test_config_and_status()` function to prioritize CSV data over text parsing
- Added new `extract_config_from_csv()` function for reliable configuration extraction

### Fixed
- More accurate shape, dtype, and layout detection from profiler results
- Improved reliability of configuration reporting in test summaries
- Better handling of tensor dimension parsing (e.g., "32[32]" format)

### Technical
- CSV-based extraction provides structured, consistent data vs. unreliable text parsing
- Maintains backward compatibility with text parsing as fallback
- Cleaner, more maintainable codebase with reduced complexity

## [0.1.0] - 2025-07-14

### Added
- Initial release of ttperf CLI tool
- Support for profiling TT-Metal tests with pytest
- Automatic CSV path extraction from profiler output
- Device kernel duration calculation
- Real-time output streaming
- Flexible command-line argument parsing
- Support for named profiles
- Comprehensive error handling

### Features
- Simple CLI interface: `ttperf [name] [pytest] <test_path>`
- Automatic detection of test files and paths
- Integration with TT-Metal profiler tools
- CSV parsing for performance metrics
- Real-time progress monitoring

### Dependencies
- pandas for CSV processing
- Python 3.7+ support
- TT-Metal development environment

## [Unreleased]

### Planned
- Enhanced error messages

##