v1.15.0: ROCMExecutionProvider support

ROCMExecutionProvider support

The Optimum ONNX Runtime integration is extended to officially support ROCMExecutionProvider. See more details in the documentation.

• Add AMD GPU support by @mht-sharma in #1546
• Update ROCM ORT doc by @mht-sharma in #1564

Extended ONNX export

The Swin2sr, DPT, GLPN, ConvNextv2 are now supported in the ONNX export.

• Swin2sr onnx by @baskrahmer in #1492
• Add depth-estimation w/ DPT+GLPN by @xenova in #1529
• Add convnextv2 onnx export by @xenova in #1560

What's Changed

• Add OV export CLI to README by @echarlaix in #1526
• Refactor NormalizedConfigs for GQA by @michaelbenayoun in #1539
• Fix model patcher ONNX decoder export by @fxmarty in #1547
• Add AMD to the documentation main page by @mfuntowicz in #1540
• Add Optimum-amd documentation to the PR & release doc by @fxmarty in #1562
• Add amd documentation by @echarlaix in #1557
• Remove delete_doc_comment workflows by @regisss in #1565
• optimum-nvidia by @mfuntowicz in #1566
• Update installation instructions in README by @echarlaix in #1568
• Update doc for AMD by @mht-sharma in #1570
• Add amd extra to setup.py by @echarlaix in #1567

New Contributors

• @xenova made their first contribution in #1529

Full Changelog: v1.14.0...v1.15.0

v1.14.1: Patch release

• Update optimum-intel required version by @echarlaix in #1521
• Swin2sr onnx by @baskrahmer in #1492
• Fix Falcon ONNX export with alibi by @fxmarty in #1524
• Fix whisper v3 ONNX export by @fxmarty in #1525
• Add new fusion argument to fix compatibility with onnxruntime v1.16.2 by @echarlaix in #1535
• Add depth-estimation w/ DPT+GLPN by @xenova in #1529

v1.14.0: LCMs, SpeechT5, Falcon, Mistral, decoder refactorization

ONNX

New architectures

Falcon

• Add ONNX and ORT support for Falcon by @fxmarty in #1391

SpeechT5

• SpeechT5 ONNX support by @fxmarty in #1404

Mistral

• Add Mistral models ONNX export support by @echarlaix in #1425

TrOCR

• Enable KV cache support by @fxmarty in #1456

LCMs

Enable LCMs (available in in diffusers since v0.22.0) ONNX export and ORT inference by @echarlaix in #1469

from optimum.onnxruntime import ORTLatentConsistencyModelPipeline

pipe = ORTLatentConsistencyModelPipeline.from_pretrained("SimianLuo/LCM_Dreamshaper_v7", export=True)
prompt = "sailing ship in storm by Leonardo da Vinci"
images = pipe(prompt=prompt, num_inference_steps=4, guidance_scale=8.0).images

Also enable ONNX export using the CLI :

optimum-cli export onnx --model SimianLuo/LCM_Dreamshaper_v7 lcm_onnx/

Decoder refactorization

• Add position ids as input during ONNX export by @fxmarty in #1381
• Enable the export of only one decoder for decoder-only models by @echarlaix in #1257

GPTQ

• Enable possibility to choose exllamav2 kernels for GPTQ models by @SunMarc in #1419
• Disable exllamav2 for quantization by @SunMarc in #1482
• Default to exllama when exllamav2 is disabled by @SunMarc in #1494
• Added cache_block_outputs parameter to handle models with non-regular structure such as ChatGLM by @AlexKoff88 in #1479
• Add support for CPU Inference by @vivekkhandelwal1 in #1496
• Fix minimum version of auto-gptq by @fxmarty in #1504
• switch to exllama_config instead of disabling exllamav2 by @SunMarc in #1505

Other changes and bugfixes

• Fix wrong dtype in the ONNX export by @fxmarty in #1369
• Add support for loading quantization from config by @aarnphm #1363
• Guard multiprocessing set start method by @fxmarty in #1377
• Do not output KV cache when not using with-past in the ONNX export by @fxmarty in #1358
• Fix provider availability check on ORT 1.16.0 release by @fxmarty in #1403
• Fix quantization for onnxruntime v1.16.0 by @echarlaix in #1405
• Fix normalized config key for models architecture by @echarlaix in #1408
• Fix arg in bettertransformer llama attention by @SunMarc in #1421
• Ignore .xml files for Stable Diffusion ORT downloads by @baskrahmer in #1428
• Falcon BetterTransformer requires transformers>=4.34 by @fxmarty in #1431
• Fix llama ONNX export by @fxmarty in #1432
• Update attention.py by @DongHande in #1416
• Remove SharedDDP as it was deprecated from Transformers by @AdamLouly in #1443
• Fix owlvit task detection by @fxmarty in #1453
• Improve ONNX quantization doc by @fxmarty in #1451
• Fix perceiver tests and dummy inputs for ONNX by @baskrahmer in #1449
• Disable bart onnx export for text-classification and question-answering by @fxmarty in #1457
• Fix ONNX exporter library_name by @baskrahmer in #1460
• [ORT Training] Some important updates of ONNX Runtime training APIs by @JingyaHuang in #1335
• Fix typo in BetterTransformer CLIP by @fxmarty in #1468
• Fix custom architecture detection in onnx export by @fxmarty in #1472
• Fix whisper export by @mht-sharma in #1503
• Update Transformers dependency for Habana extra by @regisss in #1508
• Fix argument error by @ranchlai in #1501
• Remove attention mask patching by @fxmarty in #1509
• Fix generation input by @echarlaix in #1512
• Fix tests ORTModel by @fxmarty in #1517
• Fix BT on transformers 4.35 release by @fxmarty in #1518

New Contributors

• @aarnphm made their first contribution in #1363
• @DongHande made their first contribution in #1416
• @AlexKoff88 made their first contribution in #1479
• @vivekkhandelwal1 made their first contribution in #1496
• @ranchlai made their first contribution in #1501

v1.13.3: Patch release

Patch release for transformers==4.34.1 compatibility. We will do a release next week for transformers==4.35 compatibility and new features. Please bear with us!

• Falcon BetterTransformer requires transformers>=4.34 by @fxmarty #1431
• Fix arg in bettertransformer llama attention by @SunMarc #1421
• Update Transformers dependency for Habana extra by @regisss #1508
• temporarily pin to transformers<4.35 by @fxmarty 6169310

v1.13.2: Patch release

• Fix provider availability check on ORT 1.16.0 release by @fxmarty in #1403
• Fix ONNX Runtime quantization compatibility for onnxruntime v1.16.0 by @echarlaix in #1405

v1.13.1: Patch release

Fix ONNX fp16 export that broke in 1.13.0.

What's Changed

• Fix wrong dtype in the ONNX export by @fxmarty in #1369
• Fix tests collection for TFLite export and trigger TFLite tests only when relevant by @fxmarty in #1368
• upgrade min compatible optimum-intel version by @echarlaix in #1371
• Fix fp16 ONNX export test by @fxmarty in #1373

v1.13.0: ONNX weight deduplication, ONNX export and ORT extension

Deduplicate Embedding / LM head weight in the ONNX export

Workaround for a bug in the PyTorch ONNX export that does not deduplicate the Embedding and LM head shared weight: pytorch/pytorch#108342. For small enough models, this results in up to 50% ONNX serialized model size decrease.

• Fix PyTorch tied weights being duplicated in the exported ONNX models by @fxmarty in #1326
• Fix initializer detection for weight deduplication by @fxmarty in #1333

Extended ONNX Runtime support

ONNX Runtime integration now supports Pix2Struct and MPT architectures. Donut now supports IO Binding. Encoder-Decoder models are now supported as well.

• Pix2Struct onnxruntime support by @krathul in #1296
• Add MPT onnx and ORT support by @jiqing-feng in #1161
• Donut iobinding by @IlyasMoutawwakil in #1209
• Add encoder decoder model by @mht-sharma in #851

Extended ONNX export: MPT, TIMM models, Encoder-Decoder

Additionally, the model SAM is now be default exported as a vision_encoder.onnx, and prompt_encoder_mask_decoder.onnx.

• Add MPT onnx and ORT support by @jiqing-feng in #1161
• Adds ONNX Export Support for Timm Models by @mht-sharma in #965
• Add encoder decoder model by @mht-sharma in #851
• Fix SAM ONNX export requirements with transformers 4.32, export vision encoder separately by @fxmarty in #1301

BetterTransformer supports Falcon

• [BetterTransformer] Add falcon to BetterTransformer by @younesbelkada in #1343

Major bugfix: ability to set GPTQ Exllama kernel maximum length in the transformers integration

The function exllama_set_max_input_length from auto-gptq can now be used with Transformers GPTQ models.

• Version bump + add max_input_length to gptq by @SunMarc in #1329

Other changes and bugfixes

• Update version to 1.12.1.dev0 following release by @fxmarty in #1312

• Add GPTQ prefill benchmark by @fxmarty in #1313

• Precise ORTModel documentation by @fxmarty in #1268

• Improve BetterTransformer backward compatibility by @fxmarty in #1314

• Improve ORTModel documentation by @fxmarty in #1245

• Add bitsandbytes benchmark by @fxmarty in #1320

• fix typo in log message by @AAnirudh07 in #1322

• Support customize dtype for dummy generators by @JingyaHuang in #1307

• Fix opset custom onnx export by @mht-sharma in #1331

• Replace mpt to ernie custom export by @mht-sharma in #1332

• Fix BT benchmark script by @fxmarty in #1344

• Add name_or_path for donut generation by @fxmarty in #1345

• send both negative prompt embeds to ORT SDXL by @ssube in #1339

• add vae image processor by @echarlaix in #1219

• add negative prompt test by @echarlaix in #1347

• Add GPT BigCode to the BT documentation by @fxmarty in #1356

• Add BT dummy objects by @fxmarty in #1355

• Add text2text-generation-with-past test for encoder-decoder model by @mht-sharma in #1338

• Fix sentence transformer export by @mht-sharma in #1366

New Contributors

• @krathul made their first contribution in #1296
• @AAnirudh07 made their first contribution in #1322
• @jiqing-feng made their first contribution in #1161
• @ssube made their first contribution in #1339

Full Changelog: v1.12.0...v1.13.0

v1.12.0: AutoGPTQ integration, extended BetterTransformer support

AutoGPTQ integration

Part of AutoGPTQ library has been integrated in Optimum, with utilities to ease the integration in other Hugging Face libraries. Reference: https://huggingface.co/docs/optimum/llm_quantization/usage_guides/quantization

• Add GPTQ Quantization by @SunMarc in #1216
• Fix GPTQ doc by @regisss in #1267
• Add AutoGPTQ benchmark by @fxmarty in #1292
• Fix gptq params by @SunMarc in #1284

Extended BetterTransformer support

BetterTransformer now supports BLOOM and GPT-BigCode architectures.

• Bt bloom by @baskrahmer in #1221
• Support gpt_bigcode in bettertransformer by @fxmarty in #1252
• Fix BetterTransformer starcoder init by @fxmarty in #1254
• Fix BT starcoder fp16 by @fxmarty in #1255
• SDPA dispatches to flash for MQA by @fxmarty in #1259
• Check output_attentions is False in BetterTransformer by @fxmarty in #1306

Other changes and bugfixes

• Update bug report template by @fxmarty in #1266
• Fix ORTModule uses fp32 model issue by @jingyanwangms in #1264
• Fix build PR doc workflow by @fxmarty in #1270
• Avoid triggering stop job on label by @fxmarty in #1274
• Update version following 1.11.1 patch by @fxmarty in #1275
• Fix fp16 ONNX detection for decoder models by @fxmarty in #1276
• Update version following 1.11.2 patch by @regisss in #1291
• Pin tensorflow<=2.12.1 by @fxmarty in #1305
• ONNX: disable text-generation models for sequence classification & fixes for transformers 4.32 by @fxmarty in #1308
• Fix staging tests following transformers 4.32 release by @fxmarty in #1309
• More fixes following transformers 4.32 release by @fxmarty in #1311

New Contributors

• @SunMarc made their first contribution in #1216
• @jingyanwangms made their first contribution in #1264

Full Changelog: v1.11.2...v1.12.0

v1.11.2: Patch release

Remove the Transformers version constraint on optimum[habana].

• Remove Transformers version constraint on Optimum Habana #1290 by @regisss

Full Changelog: v1.11.1...v1.11.2

v1.11.1: Patch release

Minor fix: documentation building for 1.11.

• Accelerate as a soft dependency by @fxmarty

Full Changelog: v1.11.0...v1.11.1