GitLab

This change uses the updated Marlin MoE kernel from vLLM to support
MoE with activation sorting and groups.

* feat: support phi3.5 moe model loading

* fix: prefer llama base model and improve rotary logic

* feat: return reasonable generation and add integration test

* fix: run lint and update docs

* fix: rerun lint for openapi docs

* fix: prefer do_sample false unless temp is set by user, and update chat tests

* fix: small typo adjustments

* fix: consolidate long rope paths

* fix: revert greedy by default and test changes

* Vendor configuration so that we don't have to `trust_remote_code`

* Use SparseMoELayer

* Add support for dense MoE

* Some type annotations

* Add the usual model tests

* Ruff.

---------
Co-authored-by: Daniël de Kok <me@danieldk.eu>
Co-authored-by: Nicolas Patry <patry.nicolas@protonmail.com>

This change add support for MoE models that use GPTQ quantization.
Currently only models with the following properties are supported:

- No `desc_act` with tensor parallelism, unless `group_size=-1`.
- No asymmetric quantization.
- No AWQ.

* style

* update torch

* ix issues

* fix clone

* revert mkl

* added custom PA

* style

* fix style

* style

* hide env vart

* fix mixtral model

* add skinny kernel and merge fixes

* fixed style

* fix issue for sliding window models

* addressed review comments

* fix import

* improved error messag

* updated default value

* remove import

* fix imports after rebase

* float16 dep

* improve dockerfile

* cleaned dockerfile

Remove compute capability lock

We are only calling the `get_cuda_capability` function once, so avoiding
the cost of multiple calls is not really necessary yet.

* Improve support for GPUs with capability < 8

- For models that cannot use flashinfer, use flash-attn v1 + paged
  attention for models with a compute capability older than 8.
- Disable prefix caching when using paged attention.
- When using flash-attn v1, pass the key/value, rather than the
  cache, since v1 cannot use block tables.

* nix: add flash-attn-v1 to the server environment

* Move disabling prefix caching into the block of exceptions

* Capability as `usize`s

* Fix `cargo build --features google`

* Add `cargo test --features google`

* Add LoRA adapters support for Gemma2

* Make `black` formatting happy

specify how to call local adapters

* More tensor cores.

* Fixing the logic.

* Gemma is modified by this.

* Cleanup Vertex + Chat

* logprobs defaults to false.

* Parameters are optional

* Fix  docs.

* Changing back this logprobs default.

* Fixup doc.

* Let's debug that.

* Not unstable.

* Updating Cargo ?

* Wat?

* Dummy change.

* Trying some other install.

* Trying smething.

* Revert everything.

* Update Cargo lock.

* Fixing the pre-commit after rebase.

* chore: adding note for private models in quicktour doc

* Update docs/source/quicktour.md
Co-authored-by: Omar Sanseviero <osanseviero@gmail.com>

* Update docs/source/quicktour.md
Co-authored-by: vb <vaibhavs10@gmail.com>

* Update docs/source/quicktour.md
Co-authored-by: vb <vaibhavs10@gmail.com>

---------
Co-authored-by: Omar Sanseviero <osanseviero@gmail.com>
Co-authored-by: vb <vaibhavs10@gmail.com>

This replaces the custom layers in both models.

* Add support for scalar FP8 weight scales

* Support LLM compressor FP8 checkpoints on H100

On H100, we use fbgemm-gpu, which requires bfloat16 as the input dtype.
However, we wouldn't pick up fp8 quantization for models quantized with
LLM compressor. This change adds enough parsing to detect if models have
FP8-quantized weights.

* Remove stray debug print

* wip

* added v2

* Preparing for release.

* Upgrade version in docs.

* fix: wrap python basic logs in debug assertion in launcher

* use level filters instead

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

* Update to moe-kenels 0.3.1

* Attempt to fix apt failure

* Stream options.

* Fetch stuff from nix integration test for easier testing.

* Adding the assert.

* Only send the usage when asked for.

* Update the docs.

* Impure test because we need network.

* develop.

* Optional usage.

* Fixes.

* Workflow

* Move to moe-kernels package and switch to common MoE layer

This change introduces the new `moe-kernels` package:

- Add `moe-kernels` as a dependency.
- Introduce a `SparseMoELayer` module that can be used by MoE
  models.
- Port over Mixtral and Deepseek.

* Make `cargo check` pass

* Update runner

Runs the tests in a Nix build sandbox.

* Adding a test for FD.

* Fixing flashdecoding (empty batch doesn't work).

* Fixing the invalid popping.

* Fixing radix with block_size > 1

* Last reference.

* Use an actual hash.

* Update hash for slice.len() == 1

* Update the locks.

* Increasing docker timeout.

Disable by default because CI runners do not have enough GPUs.

* use ratatui not archived tui

* bump ratatui all the way with options

enable intel ipex cpu and xpu in python3.11
Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

fix: pass missing revision arg for lora adapter when loading multiple adapters

* Add nix test.

* Modifying yourself means you need to rerun.

* Fixing the test + adding click (needed for pre-commit hooks).

* Try thuis.

* Our runner + pure test (not written)

* Reemove server.

* Root user.

* Different user ?

* Add the actual test target.

* Forgot this modification.

* Add a formatter.

* Add the secrets.

* Fixed the auth token ?

* Adding the other tests.

* Missing pre-commit.

* Test requires cargo for cargo fmt.

* Update it a bit.

* Up.

* Attempting to use a cache location for the models.

* Ignore the cache for now.