diff --git a/README.md b/README.md index d487eef3d..bde4a3152 100644 --- a/README.md +++ b/README.md @@ -45,9 +45,9 @@ The `--upgrade-strategy eager` option is needed to ensure `optimum-habana` is up To use the example associated with the latest stable release, run: > ``` > git clone https://github.com/huggingface/optimum-habana -> cd optimum-habana && git checkout v1.11.0 +> cd optimum-habana && git checkout v1.11.1 > ``` -> with `v1.11.0` the version number of this release. +> with `v1.11.1` the version number of this release. ### Option 2: Use the latest main branch under development @@ -175,7 +175,7 @@ The following model architectures, tasks and device distributions have been vali | GPT-J |
  • DeepSpeed
  • |
  • Single card
  • DeepSpeed
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | GPT-NeoX |
  • DeepSpeed
  • |
  • DeepSpeed
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | OPT | |
  • DeepSpeed
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | -| Llama 2 / CodeLlama | :heavy_check_mark: | :heavy_check_mark: |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | +| Llama 2 / CodeLlama / Llama 3 | :heavy_check_mark: | :heavy_check_mark: |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • [question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/question-answering)
  • | | StableLM | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | Falcon |
  • LoRA
  • | :heavy_check_mark: |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | CodeGen | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | @@ -183,6 +183,9 @@ The following model architectures, tasks and device distributions have been vali | Mistral | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | Phi | :heavy_check_mark: |
  • Single card
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | Mixtral | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | +| Persimmon | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | +| Qwen2 |
  • Single card
  • |
  • Single card
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | +| Gemma | :heavy_check_mark: |
  • Single card
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | T5 / Flan T5 | :heavy_check_mark: | :heavy_check_mark: |
  • [summarization](https://github.com/huggingface/optimum-habana/tree/main/examples/summarization)
  • [translation](https://github.com/huggingface/optimum-habana/tree/main/examples/translation)
  • [question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/question-answering#fine-tuning-t5-on-squad20)
  • | | BART | |
  • Single card
  • |
  • [summarization](https://github.com/huggingface/optimum-habana/tree/main/examples/summarization)
  • [translation](https://github.com/huggingface/optimum-habana/tree/main/examples/translation)
  • [question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/question-answering#fine-tuning-t5-on-squad20)
  • | | ViT | :heavy_check_mark: | :heavy_check_mark: |
  • [image classification](https://github.com/huggingface/optimum-habana/tree/main/examples/image-classification)
  • | @@ -194,7 +197,9 @@ The following model architectures, tasks and device distributions have been vali | BridgeTower | :heavy_check_mark: | :heavy_check_mark: |
  • [contrastive image-text training](https://github.com/huggingface/optimum-habana/tree/main/examples/contrastive-image-text)
  • | | ESMFold | |
  • Single card
  • |
  • [protein folding](https://github.com/huggingface/optimum-habana/tree/main/examples/protein-folding)
  • | | Blip | |
  • Single card
  • |
  • [visual question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/visual-question-answering)
  • [image to text](https://github.com/huggingface/optimum-habana/tree/main/examples/image-to-text)
  • | -| Segment Anything Model | |
  • Single card
  • |
  • [object segmentation](https://github.com/huggingface/optimum-habana/tree/main/examples/object-segementation/SegmentAnythingModel)
  • | +| OWLViT | |
  • Single card
  • |
  • [zero shot object detection](https://github.com/huggingface/optimum-habana/tree/main/examples/zero-shot-object-detection)
  • | +| Segment Anything Model | |
  • Single card
  • |
  • [object segmentation](https://github.com/huggingface/optimum-habana/tree/main/examples/object-segementation)
  • | + - Diffusers: @@ -203,7 +208,8 @@ The following model architectures, tasks and device distributions have been vali | Architecture | Training | Inference | Tasks | |------------------|:--------:|:--------------------:|:------| -| Stable Diffusion | |
  • Single card
  • |
  • [text-to-image generation](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion)
  • | +| Stable Diffusion |
  • [textual inversion](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion/training#textual-inversion)
  • [ControlNet](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion/training#controlnet-training)
  • |
  • Single card
  • |
  • [text-to-image generation](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion)
  • | +| Stable Diffusion XL |
  • [fine-tuning](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion/training#fine-tuning-for-stable-diffusion-xl)
  • |
  • Single card
  • |
  • [text-to-image generation](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion)
  • | | LDM3D | |
  • Single card
  • |
  • [text-to-image generation](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion)
  • | @@ -212,10 +218,11 @@ The following model architectures, tasks and device distributions have been vali
    -| Architecture | Training | Inference | Tasks | -|------------------|:--------:|:--------------------:|:------| -| Llama 2 | :heavy_check_mark: | |
  • [DPO Pipeline](https://github.com/huggingface/optimum-habana/tree/main/examples/trl)
  • | -| Llama 2 | :heavy_check_mark: | |
  • [PPO Pipeline](https://github.com/huggingface/optimum-habana/tree/main/examples/trl)
  • | +| Architecture | Training | Inference | Tasks | +|------------------|:--------:|:--------------------:|:-----------------------------------------------------------------------------------------------| +| Llama 2 | :heavy_check_mark: | |
  • [DPO Pipeline](https://github.com/huggingface/optimum-habana/tree/main/examples/trl)
  • | +| Llama 2 | :heavy_check_mark: | |
  • [PPO Pipeline](https://github.com/huggingface/optimum-habana/tree/main/examples/trl)
  • | +| Stable Diffusion | :heavy_check_mark: | |
  • [DDPO Pipeline](https://github.com/huggingface/optimum-habana/tree/main/examples/trl)
  • |
    diff --git a/docs/source/index.mdx b/docs/source/index.mdx index cb2552cd6..2e4d810ef 100644 --- a/docs/source/index.mdx +++ b/docs/source/index.mdx @@ -47,7 +47,7 @@ In the tables below, ✅ means single-card, multi-card and DeepSpeed have all be | GPT-J |
  • DeepSpeed
  • |
  • Single card
  • DeepSpeed
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | GPT-NeoX |
  • DeepSpeed
  • |
  • DeepSpeed
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | OPT | |
  • DeepSpeed
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | -| Llama 2 / CodeLlama | ✅ | ✅ |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | +| Llama 2 / CodeLlama / Llama 3 | ✅ | ✅ |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • [question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/question-answering)
  • | | StableLM | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | Falcon |
  • LoRA
  • | ✅ |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | CodeGen | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | @@ -55,6 +55,9 @@ In the tables below, ✅ means single-card, multi-card and DeepSpeed have all be | Mistral | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | Phi | ✅ |
  • Single card
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | Mixtral | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | +| Gemma | ✅ |
  • Single card
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | +| Qwen2 |
  • Single card
  • |
  • Single card
  • |
  • [language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | +| Persimmon | |
  • Single card
  • |
  • [text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)
  • | | T5 / Flan T5 | ✅ | ✅ |
  • [summarization](https://github.com/huggingface/optimum-habana/tree/main/examples/summarization)
  • [translation](https://github.com/huggingface/optimum-habana/tree/main/examples/translation)
  • [question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/question-answering#fine-tuning-t5-on-squad20)
  • | | BART | |
  • Single card
  • |
  • [summarization](https://github.com/huggingface/optimum-habana/tree/main/examples/summarization)
  • [translation](https://github.com/huggingface/optimum-habana/tree/main/examples/translation)
  • [question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/question-answering#fine-tuning-t5-on-squad20)
  • | | ViT | ✅ | ✅ |
  • [image classification](https://github.com/huggingface/optimum-habana/tree/main/examples/image-classification)
  • | @@ -66,14 +69,16 @@ In the tables below, ✅ means single-card, multi-card and DeepSpeed have all be | BridgeTower | ✅ | ✅ |
  • [contrastive image-text training](https://github.com/huggingface/optimum-habana/tree/main/examples/contrastive-image-text)
  • | | ESMFold | |
  • Single card
  • |
  • [protein folding](https://github.com/huggingface/optimum-habana/tree/main/examples/protein-folding)
  • | | Blip | |
  • Single card
  • |
  • [visual question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/visual-question-answering)
  • [image to text](https://github.com/huggingface/optimum-habana/tree/main/examples/image-to-text)
  • | -| SAM | |
  • Single card
  • |
  • [object segmentation](https://github.com/huggingface/optimum-habana/tree/main/examples/object-segementation/SegmentAnythingModel)
  • | +| OWLViT | |
  • Single card
  • |
  • [zero shot object detection](https://github.com/huggingface/optimum-habana/tree/main/examples/zero-shot-object-detection)
  • | +| SAM | |
  • Single card
  • |
  • [object segmentation](https://github.com/huggingface/optimum-habana/tree/main/examples/object-segementation)
  • | - Diffusers -| Architecture | Training | Inference | Tasks | -|------------------|:--------:|:---------:|:------| -| Stable Diffusion | |
  • Single card
  • |
  • [text-to-image generation](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion)
  • | -| LDM3D | |
  • Single card
  • |
  • [text-to-image generation](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion)
  • | +| Architecture | Training | Inference | Tasks | +|---------------------|:--------:|:---------:|:------| +| Stable Diffusion |
  • [textual inversion](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion/training#textual-inversion)
  • [ControlNet](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion/training#controlnet-training)
  • |
  • Single card
  • |
  • [text-to-image generation](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion)
  • | +| Stable Diffusion XL |
  • [fine-tuning](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion/training#fine-tuning-for-stable-diffusion-xl)
  • |
  • Single card
  • |
  • [text-to-image generation](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion)
  • | +| LDM3D | |
  • Single card
  • |
  • [text-to-image generation](https://github.com/huggingface/optimum-habana/tree/main/examples/stable-diffusion)
  • | - TRL: @@ -82,6 +87,7 @@ In the tables below, ✅ means single-card, multi-card and DeepSpeed have all be |------------------|:--------:|:--------------------:|:------| | Llama 2 | ✅ | |
  • [DPO Pipeline](https://github.com/huggingface/optimum-habana/tree/main/examples/trl)
  • | | Llama 2 | ✅ | |
  • [PPO Pipeline](https://github.com/huggingface/optimum-habana/tree/main/examples/trl)
  • | +| Stable Diffusion | ✅ | |
  • [DDPO Pipeline](https://github.com/huggingface/optimum-habana/tree/main/examples/trl)
  • | Other models and tasks supported by the 🤗 Transformers and 🤗 Diffusers library may also work.