Commit Graph

1059 Commits

Author SHA1 Message Date
drbh 8ad20daf33
CI (2599): Update ToolType input schema (#2601)
* Update ToolType input schema

* lint

* fix: run formatter

* fix: allow tool choide to be null

---------

Co-authored-by: Wauplin <lucainp@gmail.com>
2024-10-08 12:35:48 -04:00
Daniël de Kok 6db3bcb700
nix: move back to the tgi-nix main branch (#2620) 2024-10-08 12:55:05 +02:00
Daniël de Kok 64142489b6
Add support for fused MoE Marlin for AWQ (#2616)
* Add support for fused MoE Marlin for AWQ

This uses the updated MoE Marlin kernels from vLLM.

* Add integration test for AWQ MoE
2024-10-08 11:56:41 +02:00
Nicolas Patry 8b295aa498
Upgrade minor rust version (Fixes rust build compilation cache) (#2617)
* Upgrade minor rust version (Fixes rust build compilation cache)

* Black
2024-10-08 09:42:50 +02:00
Wang, Yi 57f9685dc3
enable mllama in intel platform (#2610)
Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
2024-10-07 21:15:09 +02:00
Florian Zimmermeister 0da4df4b96
Fix FP8 KV-cache condition (#2611)
Update kv_cache.py
2024-10-07 09:34:19 +02:00
Daniël de Kok 2358c2bb54
Add basic FP8 KV cache support (#2603)
* Add basic FP8 KV cache support

This change adds rudimentary FP8 KV cache support. The support is
enabled by passing `--kv-cache-dtype fp8_e5m2` to the launcher. Doing so
uses this type for the KV cache. However support is still limited:

* Only the `fp8_e5m2` type is supported.
* The KV cache layout is the same as `float16`/`bfloat16` (HND).
* The FP8 KV cache is only supported for FlashInfer.
* Loading of scales is not yet supported.

* Fix Cargo.toml
2024-10-04 17:51:48 +02:00
Daniël de Kok 68103079f4
nix: example of local package overrides during development (#2607) 2024-10-04 16:52:42 +02:00
drbh 3011639ff7
Revert "Unroll notify error into generate response" (#2605)
Revert "Unroll notify error into generate response (#2597)"

This reverts commit d22b0c1fbe.
2024-10-03 17:56:40 -04:00
Nicolas Patry f6e2f05b16
New release 2.3.1 (#2604)
* New release 2.3.1

* Update doc number
2024-10-03 14:43:49 +02:00
drbh d22b0c1fbe
Unroll notify error into generate response (#2597)
* feat: unroll notify_error if no tool is choosen

* fix: expect simple message when no tool is selected

* fix: improve test to avoid notify_error

* fix: improve docs and indicate change in expected response

* fix: adjust linting in test file
2024-10-02 11:34:57 -04:00
drbh 2335459556
CI (2592): Allow LoRA adapter revision in server launcher (#2602)
allow revision for lora adapters from launcher

Co-authored-by: Sida <sida@kulamind.com>
Co-authored-by: teamclouday <teamclouday@gmail.com>
2024-10-02 10:51:04 -04:00
Nicolas Patry 0204946d26
Max token capacity metric (#2595)
* adding max_token_capacity_metric

* added tgi to name of metric

* Adding max capacity metric.

* Add description for the metrics

---------

Co-authored-by: Edwinhr716 <Edandres249@gmail.com>
2024-10-02 16:32:36 +02:00
Nicolas Patry d18ed5cfc5
Mllama flash version (#2585)
* Working loading state.

* Preprocessing.

* Working state ? (Broke idefics1 temporarily).

* Cleaner condition.

* Fix idefics.

* Updating config, removing TODO

* Mllama

* Ugrade transformers 4.45

* Flashing mllama.

* Starting to get there.

* Working state.

* Integrations tests for mllama (cutting to 10 tokens because there seems'
to be instability after (meaning size of the batch matters.

* Updating model link.

* Earlier assert.

* Fix vlm ?

* remove log.

* Force ignore all images but last.

* Default dtype bfloat16.

* Update integration test after switch to bf16.

* Remove dead code.

* Removed dead code.

* Upgrade the flake to latest transformers/tokenizers

* Move to hf tgi-nix

* Upgrade to 0.5.0
2024-10-02 11:22:13 +02:00
Daniël de Kok 584b4d7a68
nix: experimental support for building a Docker container (#2470)
* nix: experimental support for building a Docker image

Run using something like:

```
docker run \
  --device nvidia.com/gpu=all \
  -it --rm -p 8080:80 \
  -v $PWD/data:/data \
  -v $PWD/tmp:/tmp \
  tgi-docker:latest \
  --model-id <model_id>
```

* Example of building the Docker image using Nix inside Docker

* Stream to make the builder image smaller

This avoids storing a Docker image tarball in the image. Instead,
stream the layers while doing `docker run`.

* Don't spam journalctl on Linux

* Other dockerfile.

---------

Co-authored-by: Nicolas Patry <patry.nicolas@protonmail.com>
2024-10-01 18:02:06 +02:00
Daniël de Kok 1c84a30fe6
MoE Marlin: support `desc_act` for `groupsize != -1` (#2590)
This change uses the updated Marlin MoE kernel from vLLM to support
MoE with activation sorting and groups.
2024-09-30 19:40:25 +02:00
Daniël de Kok d1f257ac56
Move flake back to tgi-nix `main` (#2586) 2024-09-30 11:39:41 +02:00
drbh 93a7042d7e
feat: support phi3.5 moe (#2479)
* 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>
2024-09-30 11:15:09 +02:00
Daniël de Kok 90a1d04a2f
Add support for GPTQ-quantized MoE models using MoE Marlin (#2557)
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.
2024-09-30 11:14:32 +02:00
Mohit Sharma f9e561eced
Update ROCM libs and improvements (#2579)
* 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
2024-09-30 10:54:32 +02:00
Ikram Ul Haq e790cfc0e4
Update architecture.md (#2577) 2024-09-30 08:56:20 +02:00
Daniël de Kok afc7ded84f
Remove compute capability lazy cell (#2580)
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.
2024-09-30 08:48:47 +02:00
Daniël de Kok 1028996fb3
flashinfer: pass window size and dtype (#2574) 2024-09-28 18:41:41 +02:00
Daniël de Kok 5b6b74e21d
Improve support for GPUs with capability < 8 (#2575)
* 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
2024-09-27 16:19:42 +02:00
Alvaro Bartolome 0aa66d693a
Fix build with `--features google` (#2566)
* Fix `cargo build --features google`

* Add `cargo test --features google`
2024-09-26 11:41:38 +02:00
Alvaro Bartolome 0b7df77178
Add LoRA adapters support for Gemma2 (#2567)
* Add LoRA adapters support for Gemma2

* Make `black` formatting happy
2024-09-26 10:54:08 +02:00
Nicholas Broad 7efcb5e0ed
remove LORA_ADAPTERS_PATH (#2563)
specify how to call local adapters
2024-09-25 01:20:15 +02:00
Nicolas Patry dd8691b7c5
More tensor cores. (#2558)
* More tensor cores.

* Fixing the logic.

* Gemma is modified by this.
2024-09-24 23:57:26 +02:00
Nicolas Patry c032280b17
Cleanup Vertex + Chat (#2553)
* 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.
2024-09-24 23:37:17 +02:00
Nicolas Patry 75c8c54ac9
Hotfixing main. (#2562) 2024-09-24 23:00:43 +02:00
Aritra Roy Gosthipaty e6d29656b5
Adding note for private models in quick-tour document (#2548)
* 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>
2024-09-24 15:06:53 +02:00
Orhun Parmaksız 8024ded58f
Simplify crossterm imports (#2545) 2024-09-24 14:57:20 +02:00
Orhun Parmaksız 03263f5e88
Update the link to the Ratatui organization (#2546) 2024-09-24 14:51:48 +02:00
Daniël de Kok 3f14cd1420
Add `DenseMoELayer` and wire it up in Mixtral/Deepseek V2 (#2537)
This replaces the custom layers in both models.
2024-09-24 14:27:06 +02:00
Daniël de Kok c29dc89c18
Add support for scalar FP8 weight scales (#2550)
* 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
2024-09-24 13:57:40 +02:00
Nicolas Patry 0ff6ff60ad
Hotfixing main (#2556) 2024-09-24 11:51:14 +02:00
Nicolas Patry 74d3ce106e
Micro cleanup. (#2555) 2024-09-24 11:19:24 +02:00
Alvaro Bartolome d31a6f75cc
Remove duplicated `RUN` in `Dockerfile` (#2547) 2024-09-24 10:19:13 +02:00
OlivierDehaene 10e6f29295
chore: Add old V2 backend (#2551)
* wip

* added v2
2024-09-24 08:38:17 +02:00
Daniël de Kok 9263817c71
nix: remove unused `_server.nix` file (#2538) 2024-09-23 09:43:23 +02:00
Nicolas Patry 169178b937
Preparing for release. (#2540)
* Preparing for release.

* Upgrade version in docs.
2024-09-20 17:42:04 +02:00
OlivierDehaene 7e2d18877e
fix: wrap python basic logs in debug assertion in launcher (#2539)
* fix: wrap python basic logs in debug assertion in launcher

* use level filters instead
2024-09-20 14:59:31 +00:00
Wang, Yi f478aa77ad
hotfix: ipex fails since cuda moe kernel is not supported (#2532)
Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
2024-09-20 10:02:55 +02:00
Daniël de Kok abd24dd385
doc: clarify that `--quantize` is not needed for pre-quantized models (#2536) 2024-09-19 22:17:15 +02:00
Daniël de Kok c103760172
Update to moe-kenels 0.3.1 (#2535)
* Update to moe-kenels 0.3.1

* Attempt to fix apt failure
2024-09-19 22:16:32 +02:00
Nicolas Patry f512021e77
Stream options. (#2533)
* 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
2024-09-19 20:50:37 +02:00
Daniël de Kok ce85efa968
Move to moe-kernels package and switch to common MoE layer (#2511)
* 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
2024-09-17 18:08:58 +02:00
OlivierDehaene 86984e3236
fix: metrics unbounded memory (#2528) 2024-09-17 16:01:28 +00:00
Daniël de Kok 71e4268600
nix: pure Rust check/fmt/clippy/test (#2525)
Runs the tests in a Nix build sandbox.
2024-09-17 12:14:30 +02:00
Nicolas Patry 38fcafcf96
Adding a test for FD. (#2516)
* 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.
2024-09-16 17:00:54 +02:00